Pinzani, Roberto - The Problem of Universals From Boethius To John of Salisbury PDF
Pinzani, Roberto - The Problem of Universals From Boethius To John of Salisbury PDF
Pinzani, Roberto - The Problem of Universals From Boethius To John of Salisbury PDF
Brill’s Studies in
Intellectual History
General Editor
Founded by
Arjo Vanderjagt
Editorial Board
VOLUME 282
By
Roberto Pinzani
LEIDEN | BOSTON
Cover illustration: Arbor porphyriana, or Porphyry’s Tree. Artist: Bartolomäus Strahovsky (fl. 1701–1750).
Biblioteka Narodowa, Warsaw.
Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface.
issn 0920-8607
isbn 978-90-04-37114-9 (hardback)
isbn 978-90-04-37115-6 (e-book)
Introduction 1
1 The Problem 5
1.1 Abstract Entities 5
1.2 Predicates 6
1.3 The Relation of Predication in Aristotle 8
1.4 How Many Questions? 18
9 Gilbert of Poitiers 235
9.1 The Distinction between id quod est and esse 235
9.2 Matter, Form, Nature 241
9.3 Categorization 246
9.4 Particulars 253
9.5 Mathematical Entities and Universals 261
Bibliography 299
Index 309
Introduction
The textual journey I intend to undertake in this work leads from Boethius’
commentaries on Porphyry to 12th century discussions on universals. As with
all travels, a map and a travel guide are required, as well as a consideration of
the resources, the places and the local regulations. Let me begin saying some-
thing regarding these aspects, starting with the map.
The map: the issue of universals is often compared to the issue of prop-
erties, (cf., for instance, C. Swoyer’s article ‘Properties’ for The Stanford
Encyclopedia of Philosophy) or to that involved in the type-token distinction
(cf. D.M. Armstrong, Universals. An opinioned introduction). Neither approach
is entirely suitable for addressing the problem. The concept of property, as
we shall see, does not have a privileged position in ancient and mediaeval
categorizations; properties, in the modern sense of the term, are not always
and only stored in a single drawer bearing the label ‘properties’. For example,
Aristotelian qualities may not be good candidates for the role of ‘property’;
Priscian’s qualities may be, according to Priscian, indeed, qualities are signified
by any noun, substantive or adjectival; Boethius and Abelard sometimes use
the term ‘property’ when discussing universals; Abelard – following Aristotle
and Priscian – distinguishes different senses of the terms ‘quality’ and ‘acci-
dent’. In any case, speaking of properties amounts to anticipating a certain so-
lution to Porphyry’s questions about species and genera.
As to the type-token distinction, what is usually meant by it is a relationship
between single occurrences and their supposed printing plate or model: if, for
instance, I write ‘the’, ‘the’, ‘the’, I do so supposedly on the basis of a ‘the’ model
stored somewhere in my mind. I have no knowledge of an analogous distinc-
tion in the time period considered in this book, with the possible exception
of some remarks made by John of Salisbury. Armstrong suggests that the nar-
row sense described above could actually be the same (broad) sense in which
we say, for instance, that a single man is an instance of the type ‘man’. To see
species and genera as types, in this sense, is to say too little and too much. It
is saying too much because Porphyry’s question pertains to the meaning of
Aristotle’s words. On the other hand, it is saying too little since the problem
is precisely to understand what a type is, what properties it has, what kind of
relationship links it to its tokens or instances.1
1 Actually, having established that the problem of universals pertains to the token-type dis-
tinction, Armstrong considers various ways to exemplify these concepts (D.M. Armstrong,
Universals. An Opinioned Introduction, Westiew Press, Boulder 1989; ch. 1). I still have two
After these initial warnings, I should now say something about the
‘destination’. The way in which Porphyry lays out the problem of genera and
species, provides us with the route to follow in later discussions on the issue
of universals. Porphyry reaches mediaeval philosophers through Boethius’
commentaries, known partially to Encyclopaedists of the High Middle Ages
and entirely from the end of the 11th century. The key text for this tradition of
commentators is the Isagoge, in particular the preface, where the Neoplatonic
philosopher speaks about the ontological status of genera and species. It is
important to stress that the context is that of an introduction to Categories.
Therefore, Porphyry takes for granted a certain classification of objects, rather
than considering other alternatives, which were available at the time he wrote
his book. In the Categories Aristotle classifies objects and, at the same time,
words; what he says clearly shows how important the strict correspondence
between words and things was for him. A second point to be considered is that
the focus of Porphyry’s treatise is the meaning of certain Aristotelian techni-
cal terms: ‘species’, ‘genus’, ‘difference’, ‘propium’ and ‘accident’; thus Porphyry
does not deal with things considered per se, but with respect to their (meta)
properties and relations.
The travel guide: In the first chapter I will examine the Aristotelian theory
of predication. Aristotle establishes a kind of principle of predication in the
Categories: universal terms are involved in a relation of predication if (and only
if) some facts expressed by ordinary sentences hold. This point of view con-
stitutes an interpretative key for approaching subsequent discussions on the
problem of universals. Depending on how a philosopher stands with respect
to the predication principles, his positions on the problem of species/genera
differ. However, the target sometimes seems to be beyond the means: even
if I consider the Aristotelian principles as a means for approaching the issue
of universals and try to highlight the different positions regarding it, there is
not always a clear formulation of such principles nor a correlation between
the predication issue and that of universals. Besides, I will be concerned with
partially eccentric metaphysical discussions that do not pertain directly to the
theory of predication or Porphyry. In spite of these different scenarios, what
I would like to suggest is that the predication issue works in the background
reservations: the first is that the very different senses in which we can understand ‘type’ do
not necessarily justify the use of the same general term (see the article ‘Types and tokens’ by
L. Wetzel for The Stanford Encyclopedia), the second is that Porphyry raises a problem of tex-
tual exegesis. My objections would fail if by ‘type’ we intended exactly what Porphyry intends
when he speaks of genera and species, that is, interpretations of a universal term.
Introduction 3
publications and received some good advice (I owe Caterina Tarlazzi a deep
debt of gratitude in this regard), the situation now seems to me to be more
complicated than I thought. The following attributions of works to which I de-
vote a chapter or section of chapter now are actually unsafe: De generali et spe-
ciali statu rerum universalium, Editio super Porphyrium, Glossae in Categorias
(in the so called Introductiones Parvulorum), Logica nostrorum petitioni socio-
rum. Fortunately, my work has no philological ambitions, but merely tries to
follow the historical-theoretical path outlined above. I am primarily interest-
ed in considering philosophical positions – whoever has endorsed them – as
stages in my textual journey. Therefore, in the case of doubt – though recalling
in footnotes the problems of attribution – I will propose adopting a kind of
convention: ‘the theory of x’ = ‘the theory attributed with some reason, but
not certainly to x’. I hope to be forgiven for this by the scrupulous exegetes of
textual problems.
Finally, we have a map showing the main textual venues, i.e., Boethius’ com-
mentaries on Porphyry and those by 12th-century philosophers; we have estab-
lished a route through different texts, in which Aristotle’s predication theory is
considered. Furthermore, there are reasons to justify the insertion of different
positions or deviations, in particular the following two: the first is the necessity
to consider background elements; the second reason is that the debate some-
times tends to exceed the boundaries set by Porphyry and makes it necessary
to account for different – but in some way related – metaphysical issues.
chapter 1
The Problem
each category, and to identify the relations that tie the different catego-
ries together; and by doing this, the metaphysician supposedly provides
us with a map of the structure of all that there is.2
1.2 Predicates
By the gods, Meno, be generous, and tell me what you say that virtue is.
There will be no difficulty, Socrates, in answering your question. Let us
take first the virtue of a man.4
2 M. J. Loux, Metaphysics, p. 4.
3 A. de Libera, in a rather picturesque way, observes that ‘le Ménon est, pour utiliser une ex-
pression médiévale, le miroir historial de la problématique des universaux. Qui se penche sur
lui voit y apparaître le problème dans sa pureté native’ (La querelle des universaux de Platon
à la fin du Moyen Age, Éditions du Seuil, Paris 1996, 51).
4 Plato, Meno, ed. I. Burnet, Clarendon, Oxford 1903; 71, d; transl. B. Jowett, at classics.mit.edu/
Plato/meno.html.
The Problem 7
1. x, y, z are f
2. what is F?
3. x, y, z are f iff x, y, z have some relationship with F
We would seem to have found, then, that the many conventions of the
many about the fair and honorable and other things are tumbled about
in the mid-region between that which is not and that which is in the true
and absolute sense. […] what of those who contemplate the very things
themselves in each case, ever remaining the same and unchanged, shall
we not say that they know and do not merely opine?6
Well, tell me (Socrates), do you think that, as you say, there are ideas,
and that these other things which partake of them are named from them,
as, for instance, those that partake of likeness become like, those that
partake of greatness great, those that partake of beauty and justice just
and beautiful?’7
The first text establishes a relationship between empirical knowledge and its
theoretical justification. Mid-region opinions (for instance ‘x is wise’) need a
justification based on the knowledge of the ‘true’ condition of things. The sec-
ond text is quite clear on this aspect. What Plato is trying to do can be seen as
an attempt to justify assertive sentences.
5 F stands for an abstract noun derived from the predicate f; for instance ‘justice’ or ‘wisdom’,
from ‘is just’ or ‘is wise’.
6 Plato, The Republic, ed. I. Burnet, Clarendon, Oxford 1903; 479 d – 480 e. Transl. P. Shorey, in
Plato, in Twelve Volumes, V (Loeb Classical Library), London 1978, p.533.
7 Parmenides, ed. I. Burnet, Clarendon, Oxford 1901; 130 e – 131 a; transl. H.N. Fowler, in Plato,
op. cit., IV, London 1977, p. 213.
8 chapter 1
There is a rather subtle point. Obviously, one can imagine Socrates question-
ing a passer-by who is inclined to concede that certain things are similar. The
situation is unnatural, even at this stage, indeed assertive sentences are rather
rare, I mean assertive sentences asserted as such. It is quite rare for someone –
jumping, so to speak, onto a dais – to state something like ‘these two apples
are similar’, and be ready to give proof of that. However artificial the situation,
when it occurs, there are two possibilities. The first is that nothing happens,
that is, the assertion is accepted or contradicted on the grounds of more or
less relevant considerations. The second possibility (important for Plato and, I
would say, essential for a couple of millennia of philosophical speculation) is
that one asks for a reason to justify the assertion.
The ‘reason why’ is not obtained by a simple empirical test, as: Socrates’ doc-
tor states that Socrates has bad digestion and requires medical examinations.
Sentences like: ‘these two apples are similar’ or ‘the shirt is red’ do not require
special empirical tests. Plato would say that one (Socrates) must certainly con-
cede that these two apples are similar and the shirt is red; however, he should
not consider these empirical facts as self-standing, but as something that has
to be theoretically justified.
Aristotle clarifies what it is meant by a thing belonging (or not) to the totality
of another thing, and what it is meant by ‘being predicated of all’:
The two perspectives referred to in the text correspond to the analytic format
of a categorical sentence with the subject term S in the first position (S is in-
cluded in P) and in the second position (P is predicated of S). Aristotle says
something more: P is predicated of the totality of S if (and only if) it is not
possible to find anything of which S is said and P is not. The word ‘anything’
refers to (the meaning of) nominal clauses suitable for functioning as subjects
in sentences in which the predicate term is taken distributively. The examples
of a syllogistic protasis supplied by Aristotle clarify the point; we consider the
following two:
It seems likely that the use of yparchein with a noun or neuter adjec-
tive as subject is an unconscious invention of Aristotle’s, made in order
10 W.C. Kneale – M. Kneale, The Development of Logic, Clarendon Press, Oxford, 1962; 62–63.
The Problem 11
of predication (possibly different from the analytical one) and its range of ap-
plication. Let us consider two key passages:
Of things which are: (a) some are said of a subject but are not in any sub-
ject. For example, man is said of a subject, the individual man, but is not
in any subject. (b) Some are in a subject but are not said of any subject.
(c) Some are both said of a subject and in a subject. (d) Some are neither
in a subject nor said of a subject.11
Whenever one thing is predicated of another as of a subject, all things
said of what is predicated will be said of the subject also. For example,
man is predicated of the individual man, and animal of man; so animal
will be predicated of the individual man also – for the individual man is
both a man and an animal.12
We therefore have a relation between things that are exclusively said of, and
one between things that are said of and at the same time are in. Aristotle’s
examples are: ‘animal is said of man’, ‘science is said of grammar’. In the for-
mer text Boethius uses the phrase ‘dicitur de subjecto’, which corresponds to
the Greek ‘kath’ypokeimenon tinos legetai’; in the latter text he writes ‘de al-
tero praedicatur ut de subiecto’, corresponding to ‘kath’eteron kategoretai os
11
Categories 1a20, Transl. J.L. Ackrill (in Aristotle, Categories and De Interpretatione, Oxford
University Press, Oxford 1975). Boethius’ translation, In Categorias Aristotelis, in PL 64
(= In Cat.169 B) – 173 C: ‘eorum quae sunt alia de subiecto quodam dicuntur, in subi-
ecto vero nullo sunt, ut homo de subiecto quidem dicitur aliquo homine, in subiecto vero
nullo est. alia autem in subiecto quidem sunt, de subiecto vero nullo dicuntur … alia vero
et de subiecto dicuntur et in subiecto sunt … alia vero neque in subiecto sunt neque de
subiecto dicuntur’.
12
Categories 1b10, transl. Ackrill. Boethius’ translation, In Cat.175 C–D: ‘quando alterum de
altero praedicatur ut de subiecto, quaecumque de eo quod praedicatur dicuntur, omnia
etiam de subiecto dicentur, ut homo de quodam homine praedicatur, animal vero de ho-
mine, ergo et de quodam homine animal praedicabitur; quidam enim homo et homo
est et animal.’ Modern translations (Categories, De l’interprétation, traduction nouvelle et
notes par J. Tricot, Paris, Vrin, 1966 ; Aristotle’s Categories and De interpretatione, trans-
lated with notes by J.L. Ackrill, Oxford, Clarendon Press, 1963; Le categorie, introduzione,
traduzione e commento di Marcello Zanatta, Milano, Rizzoli, 1989) all agree with Boethius,
except for the translation by G. Colli: ‘uomo viene predicato di un determinato uomo, e
d’altro canto la nozione di animale è predicata della nozione di uomo …’ (p. 7). I. Angelelli
(Studies on Gottlob Frege and Traditional Philosophy, Reidel 1965) quotes Cat. 2A, 19–27, in
defining the predication relation (which he calls KU). In that passage Aristotle says that
the name of the species, together with the definition, is predicated of an individual under
the species (p. 13 e n. 21).
12 chapter 1
B ℘ A, X ℘ B → X ℘ A
T: X ℘ B, B ℘ A → X ℘ A
Man (B) ℘ the individual man (A), animal (X) ℘ man (B) → animal (X) ℘
the individual man (A)
The examples do not seem to have been chosen by chance. Indeed for the
things that are said-of and are-in, Aristotle does not take the whiteness as an
example with regard to the white thing (thinking of sentences like ‘Socrates is
white’), but does take into account science with regard to grammar (thinking
of ‘grammar is a science’). In doing so, Aristotle focuses on an order relation
within a certain category, believing he is thereby safeguarding the transitivity
of the predication nexus.
What Aristotle says (and the examples he gives) lead one to think that the
domain of the predication relation must contain only things of a certain sort.
If transitivity is simply assumed, the problem can be to find models which sat-
isfy the categorial properties considered by Aristotle. In the post-Aristotelian
era things become more complicated, since more systems of categorial organi-
zation are present.
An additional problem is that of the possibly different kinds of relations,
indeed Aristotle seems to assume some kind of modal decoration for the predi-
cate ‘be said of’. Boethius presents the issue in these terms:
those things which have to do with the essence and definition are said to
be predicated of a subject.13
13 Boethius, In Cat., 175D–176A: ‘praedicatio, quae est: ‘Homo albus est’ secundum accidens
est: namque accidens quod est album de subiecto homine praedicatur sed non in eo
quod quid sit, nam cum album sit accidens, homo substantia, accidens de substantia in
eo quod quid sit praedicari non potest; ergo ista praedicatio secundum accidens dicitur.
De subiecto uero praedicari est, quoties altera res de altera in ipsa substantia praedicatur,
ut animal de homine; nam quoniam animal et substantia est et genus hominis, idcirco in
eo quod quid sit de homine praedicatur. Quare illa sola de subiecto praedicari dicuntur
quaecumque in cuiuslibet rei substantia et in definitione ponuntur.’ I shall use the brack-
eted expression /…/ to indicate any interpretation of categorial entities.
14 Grice and Code propose a curious sort of terminology to label essential predication
(izzing) and accidental predication (hazzing). Cf. H.P. Grice, ‘Aristotle on the multiplic-
ity of being’, Pacific Philosophical Quarterly, 69 (1988), 175–200; A. Code, ‘Aristotle: es-
sence and accident’, in R. Grandy – R. Warner (edd.) Philosophical Grounds of Rationality:
Intentions, Categories, Ends, Oxford University Press, Oxford 1983, 411–439; S. Marc Cohen,
‘Grice and Code on IZZing and HAZZing’: downloading from the web: http://faculty.
washington.edu/smcohen/433/GriceCode.pdf. The article is on 2008.
15 For instance, if /man/ (the collection of all men, the form ‘man’ …) is a species and
Socrates is a man, then Socrates is a species. Boethius considers the point in his comment
on the Aristotelian passages quoted: ‘Sed fortasse quisquam dicat minime uerum esse
quod dictum est, nam cum homo de Socrate praedicetur (Socrates enim homo est), de
homine uero species (homo enim species est), Socrates species esse non dicitur. […] nunc
autem species de homine non in eo quod quid sit praedicatur’ (Boethius, In Cat. 176C–D).
All interpreters of Aristotle are aware of this kind of problem; see, for ancient commen-
tators, M. Mignucci, ‘Puzzles about identity. Aristotle and his Greek commentators’, in
P. Moreaux – J. Wiesner (edd.), Aristoteles Werk und Wirkung. Erster band. Aristoteles
und seine Schule, De Gruyter, Berlin – New York 1985, p. 94 footnote 50. An example
from the 12th century is a commentary on Aristotle’s Categories in MS Padova, Biblioteca
Universitaria/University Library,- 2087 studied by F. Bottin. The author discusses the para-
doxical conclusion, and suggests some solutions. Cf. F. Bottin, ‘Quelques discussions sur
la transitivité de la predication dans l’ècole d’Alberic du Mont’, in J. Jolivet – A. De Libera
(edd.), Gilbert de Poitiers et ses contemporains. Aux origines de la Logica Modernorum,
Bibliopolis, Napoli 1987.
14 chapter 1
It is clear from what has been said that if something is said of a subject
both its name and its definition are necessarily predicated of the subject.
For example, man is said of a subject, the individual man, and the noun is
of course predicated (since you will be predicating man of the individual
man), and also the definition of man will be predicated of the individual
man (since the individual man is also a man). Thus, both the noun and
the definition will be predicated of the subject.16
The text states that, for any general naming expression X, there is a predicate
name x occurring in a (true) sentence on the predicate side. Aristotle’s exam-
ple suggests an interchange between what is predicated and its nominalized
form (in my reading, the name of a second substance).17 What it is suggested
here is a kind of abstraction or predication principle
P2 can be gathered from many Aristotelian texts, once the categorial relation
of predication is interpreted in analytical terms (‘A is said of each B’, etc.). For
example, the laws of the logical square (De Interpretatione, 7, 10, 14) and con-
version (Prior Analytics, A2) are stated for statements such as ‘every (some)
A is B’ and then applied to syllogistic formulas of the type ‘B is said of every
(some) A’. The star formulation on the right is introduced to avoid possible
mismatches between intensional interpretations of the clause on the left and
extensional ones of the clauses on the right. Intensional readings of the clause
on the left require ‘raising’ the interpretation of the sentence on the right. For
16 Categories 2a19, trans. Ackrill; Boethius’ transl.: In Cat.184 C: ‘Alia autem omnia aut de
subjectis dicuntur principalibus substantiis, aut in subjectis eisdem sunt … animal de
homine praedicatur ergo de aliquo homine, nam si de nullo … neque omnino de homine’.
17 I translate the Latin expressions ‘prima/ae substantia/ae’, ‘secunda/ae substantia/ae’
as ‘first substance/s’, ‘second substance/s’ (instead of in the more traditional way as
‘primary-secondary substance/s’) to give importance to the order relation that structures
the category of substance.
The Problem 15
Here are some example of how this can be (has been) done, for ℘″:
1 2 3 4 5
a Certainly, the transformation of nouns into terms by the adjunction of articles is precluded
to Medievals. However, logicians have been interested in the different contextual uses of ex-
pressions since the early 12th century. According to the translatio theory and later, suppositio
theory, the context determines a transformation of the noun’s meaning. For instance, in the
case of suppositio simplex (translatio ad speciem) the common noun stands for the species. It
would be appropriate to consider – a transformation from nominal expressions like ‘man’ to
terms like ‘the species man’, or something of this kind.
Terms like ‘man’, ‘animal’ can be understood as proper nouns (first row in the
table above). Medieval authors do not consider ‘homo’ in the subject position
as always standing elliptically for phrases like ‘omnis homo’, ‘quidam homo’,
etc., but in some cases (eg. in the definition) as the proper noun of an abstract
The Problem 17
18 E.M. Barth, The Logic of the Articles in Traditional Philosophy. A Contribution to the Study of
Conceptual Structures, Reidel, Dordrecht 1974, 38–39: ‘A proposition (a) the/an M is p will
be called ‘a logophoric proposition’ and will be said to express a logophoric judgment iff
the article is not used anaphorically and the proponent of (a) does not, by this proposi-
tion (a), mean: (b) the only entity in the domain possessing the property M, also has the
property P; nor does he use (a) as a stylistic variant of the proposition (c) all M’s are P, or
every M is P; nor of (c’) an arbitrary M is P; nor of (c’’) the class (or set) of M’s is included in
the class (set) of P’s; nor of (d) some M (M’s) is (are) P; and if, furthermore, the proposition
the/an M is M, in which the/an M has the same sense as in (a) and ‘is’ is used as a copula
of predication, is regarded by the proponent as (tauto)logically true.
19 Modern literature on this issue is abundant. As the authors of a recent article say ‘the his-
tory of philosophy has not been kind to arbitrary objects’ (F.J. Pelletier – N. Asher, ‘Generic
and defaults’ in J. van Benthem – A. ter Meulen, Handbook of Logic and Language,
Elesevier Amsterdam1997, p. 1139). However, there are some authors to be considered
when interpreting the terms used generically as proper nouns. See, for instance, T. Parsons
on non-existent objects (T. Parson, Nonexistent Objects, Yale U.P., New Haven – London,
1980), K. Fine on arbitrary objects (K. Fine, Reasoning with Arbitrary Objects, Blackwell,
Oxford 1985), E.N. Zalta on objects encoding properties (the Principia Metaphysica offer a
formal version of Zalta’s theory: mally.stanford.edu/principia.pdf).
18 chapter 1
species, differences, propria and accidents are. These are technical terms refer-
ring to things of different sort: ‘species’ and ‘genus’ refer to terms of a predi-
cation relation; ‘propria’ refers to properties owned exclusively by something;
‘accident’ expresses a property of qualities and, at the same time, a kind of
meta-category containing all things placed by Aristotle in the categories dif-
ferent from substance. About species and genera, Porphyry wonders whether
genera and species subsist and in what way.20
A second question is Porphyry’s: At least at first glance, Porphyry’s problem
does not regard properties, or types of objects, but genera and species, that is
to say, non-singular objects ordered according to a certain relation of predica-
tion. Whoever asks if genera and species exist may have two different things
in mind. The first one (Q1) is whether there exists, or more weakly, there is a
set of entities on which the relation of predication can be defined. The second
one (Q2) is if, for some entities ax, by and relation R, the ordinary predication
(every)(some)(a) y is an x, is explained – under certain conditions – depending
on the fact that R(ax, by). Q1 – according to Porphyry – splits up into different
problems: (Q1a) species and genera exist in reality or they are only bare con-
cepts, (Q1b) they are concrete or abstract things of some sort (Q1c) they are
separate or not separated from sensible objects. The Isagoge does not suggest
anything about Q2, this question remains in the background.
Q1 actually seems to me a unique problem, namely to establish an onto-
logical border, a domain of objects to define the predication relation. Ontology
in some way constitutes a necessary preliminary to metaphysics, providing
the pawns for playing different metaphysical games. Scholars in the twelfth
century utilize the catalogues supplied by Aristotle-Boethius and, at least par-
tially, those of the Neoplatonic tradition (see the chapter on Scotus Eriugena).
However, one cannot say that the catalogues are complementary or that the
same things are catalogued once only, under the same label. On the contrary,
objects such as forms, essences and material components can be considered
from points of view that are different and not easily comparable.
The main classification system used by medieval philosophers is the
Aristotelian one. This system reveals some anomalies, depending on the dif-
ferent classification criteria and on the fact that some objects are catalogued
more than once. Boethius in his logical commentaries adds some categories
or super-categories that further complicate the situation, or, so to speak, make
his filing cabinet even more untidy. Besides, Boethius presents different clas-
sification systems in De Institutione Arithmetica and Opuscola Theologica;
20 Porphyry, Isagoge, Boethius’ transl., in: In Isagogen Porphyrii Commenta (= In Isag.), ed.
S. Brandt, Tempsky – Freytag, Wien-Leipzig 1906, 159, 3–9.
20 chapter 1
the objects he speaks about in these writings can hardly be placed in some
Aristotelian folder.
A third question concerns the meanings of terms. Aristotle sometimes uses
a semantic terminology, for example, in the famous introductory section of De
Interpretatione, he speaks about the semantic relations among word, things
and concepts, or, in an equally famous passage of the Categories 4, he consid-
ers the meaning of different type-expressions. In the translation of Boethius:
‘eorum quae secundum nullam complexionem dicuntur, singulum aut substan-
tiam significat, aut quantitatem, aut qualitatem, aut …’. These texts provide a
hint for commentators to interpret what Aristotle says about categorical enti-
ties from a semantic point of view.
Focusing on meanings involves a change of perspective. Aristotle’s catego-
rization does not match those of grammar. There are, of course, narrower dis-
tinctions, for example, the very narrow (parvissima according to Boethius) into
(universal and particular) substances and accidents that may better match a
syntactic distinction. The observations we find in Boethius on the meaning
of terms lead one to consider an even narrower classification between what
nouns refer to and what they mean. This distinction – also present in the tra-
dition of Latin grammar – is the basis of the (medieval) ‘dualist’ approach to
semantics.
A fourth question (=Q2) is how the world is made. When I assert a categorical
sentence such as ‘Socrates is wise’ or ‘every man is an animal’, what justifies my
assertion? The question is not (only) about the existence of the wise man, wis-
dom, or the human being, but about how these things interact to build a fact.
Not all interpretations of the predication seem to work in justifying ordinary
predication: for instance the concept of collective totality (collectio) developed
by the Medievals is poorly suited to metaphysical purposes, it does not make
much sense to say that the basis of the truth of ‘Socrates is wise’ is Socrates be-
longing to the collection of wise things. Abelard, in fact, criticizes the collectio
theory as it does not allow for a semantic interpretation of terms and does not
provide an answer to the Porphyry problem, but not because it involves some
kind of inconsistency.21
21 What I have just said may sound odd to anyone who considers the (proto-) set theoretic
solutions of the universal problem consistent with a nominalist philosophy. What I fail to
understand is the thesis of ‘class nominalism’ formulated by Amstrong: ‘the class of all the
white things forms a natural class, a class with a reasonable degree of naturalness. That is
all that can be said about what makes a white thing white’ (D.M. Armstrong, Universals.
An Opinionated Introduction, Westview Press, Boulder-London 1989, p. 18). On the other
hand I understand the antithesis: ‘it is unnatural to say that it is an electron because it is a
The Problem 21
member of the class of electrons’ (p. 28). To be honest, Armstrong does not seem to regard
this observation as antithetic to the previous one, but to the traditional way to ‘give’ a set;
he writes: ‘..and that it is natural to put the property first and the class second is some
reason to think that that is the true direction of explanation. This is bad news for any class
nominalism. If type determines class membership, then membership cannot determine
class’ (ibi). The sense of oddity is weakened if we move from abstract sets to mereological
aggregates, but does not disappear completely. Anyway, it is not my task here to consider
the modern metaphysical interpretation of class, or to take up a position on it, but only to
provide documentary evidence – in the period I am concerned with – on different forms
of class nominalism.
chapter 2
To address the issue of predication from a logical point of view, one must
begin by looking at the grammar of terms that act as subjects and predi-
cates in standard categorical sentences. We find several descriptions of the
semantic categories of expressions that form sentences. The interesting thing
is that these categories do not correspond to those of Aristotle’s broad division
into ten classes. Grammatical analysis of nominal expressions and predicates
(be + adjectives, nouns, or participles) somehow becomes interwoven with
the ontological discussion, evoking several signified entities. Once the mean-
ings (or semantic categories) of the terms have been identified, a reader of
De Interpretatione can (or must) deal with truth conditions, thereby gaining a
particular perspective on the universal-particular relationship.
In Boethius’s commentaries, one can find several remarks about the relation-
ship that links a certain expression to its meaning. The interest of the subject
is twofold: we can gain information about an alternative categorization system
to the Aristotelian one, and, from another perspective, about the entities to
which expressions of language refer. The semantical analysis leads to a non-
standard approach, at least in Boethius’s time, to the problem of universals. In
this section I will deal with the meanings of general expressions. Just a prelimi-
nary warning as to Boethius’ terminology: in addition to the well-known merits
of having developed an appropriate logical language, stylistic effects and some
fluency in the use of technical terms should also be attributed to Boethius. We
often find a way of arguing that aims more to delineate concepts than to define
technical terms and link them to precise theoretical assumptions.
Among the expressions of semantic relations, naming (nominatio)1 has a
prominent role; it is understood mostly as a correspondence between nominal
expressions and particular things (concrete or abstract). Boethius uses the pas-
sive voices of the verb in cases like ‘quilibet equus singulariter animal nomina-
tur’, while the active voices seem to mean in a non-technical sense ‘talk about’.
Variants of the passive form are ‘nominari ad’, always referred to an expression
that is said of a certain set of individuals, and ‘nominari + ablative’: for example
‘homo individuis nominatur’2.
1 I translate Latin semantical expressions as closely as possible in this and other cases.
2 In Cat., 195, A. I interpret the ablative in ‘individuis nominatur’ as an ablative of limitation.
24 chapter 2
It happens that the soul of one who hears the name ‘Plato’ refers to a
single person and substance because of platonity.3
When I say ‘Socrates’ or ‘Plato’ or I name some individual substance, I
mean a something this.4
Indeed, when I say ‘Cicero’, I name one person and one substance.5
Common nouns too enjoy a naming relation. The specific term ‘homo’ means
individuals according to a semantic relation of naming.6
3 Commentarii in librum Aristotelis Peri hermeneias I–II, ed. C. Meiser, 2 voll., Lipsiae, 1877–1880
(=In De Int.), II, 137, 6–16: in this case Boethius use the verb ‘referre’ in place of ‘nominare’:
‘audientis animus Platonis vocabulum ad unam personam unamque particularem substan-
tiam referat’.
4 In Cat., 194, C–D: ‘… cum dico Socrates et Plato vel aliquam individuam substantiam nomino,
hoc aliquid significo.’
5 In Cat., 194, D: ‘cum enim dico Cicero, unam personam, unamque substantiam nominavi’.
Cfr. anche 186, D; 239, B; In De Int. II, 61, 32–62, 2.
6 ‘Nominare’ (naming) usually means ‘refer to’ and, in the passive, ‘be named’, but also ‘be a
noun for’.
7 In Cat., 167, D: ‘Commune genus est animal et communi nomine animalia nominantur’.
8 In Cat., 195 A: ‘Homo de individuis nominatur’.
Boethius on General Terms 25
Whatever is named is not the same thing for which it is named; As the
grammarian is not the same as the grammar for which it is named.9
The term ‘denominatio’ refers in this case to both the grammatical relation be-
tween the adjective and its abstract form (‘white – ‘whiteness’), and to the (set
of) things the adjective names.
The semantic relationship that seems to identify an abstract quality is the
determinatio. The difference, for example, in indicating a certain substance,
determines a quality.10 Common nouns as well as adjectives are in a relation
of determination (for instance ‘man’ determines man being = humanitas).
‘Determinatio’ is used in a different sense in the text:
This comment leads one to think that the determination of a term is its exten-
sion, to use a modern expression. (Aristotle and) Boethius add that there is a
determination of quality over the substance ( fit determinatio circa substantiam
qualitatis). In my view, therefore, determinatio can be greater or less for a factor
that links meaning to reference; In other words given a denotation of more or
fewer things, there is greater or lesser determination of a property of the term.
The schema of the main semantic relations is the following:12
Nominat → Things
9 In Cat., 220, B: ‘Omnis autem denominatio non est id quod est ea res de qua nominatur; ut
grammaticus non idem est quod grammatica de qua nominatus est’.
10 In Isag. II, 192, …: differentia enim ita substantiam demonstrat, ut circa substantiam qual-
itatem determinet’.
11 In Cat., 195, B–C: ‘Determinatio vero quoties ipse terminus multa concludit, major est, et
minor quoties pauciora, quocirca genus plurima colligit, species vero non tam plurima.’
12 Boethius’s terminology is not fixed; however, some suggestions can be gathered from
these texts for considering the semantic behavior of proper nouns for individuals and
properties.
26 chapter 2
a In Cat., 194, B.
b In Cat., 194–195 passim.
c In Cat., 195, A.
d In Isag. II, 234, 8–9.
e In Cat., 159–161 passWim; 164, B–C.
f In Isag. II, 242, 5–6.
g In De Int. II, 65, 8–9.
‘To signify a certain this’ (hoc aliquid) does not seem to mean the same thing as
‘to signify an individual thing or a species’, because Boethius distinguishes be-
tween naming individual things and signifying a certain this.13 The table shows
how the notion of signification is more general (or generic) than those of nomi-
natio and determinatio. Boethius and his twelfth-century commentators share
this terminological use.
Going back to the nominatio-determinatio I would now like to draw atten-
tion to some texts concerning proper and common nouns, where explicit men-
tion (of the kind) of their signified properties is made.
13
In Cat., 194, C–D: ‘Quemadmodum quantitas quantum significat, et qualitas quale, sic
etiam substantia videtur hoc aliquid significare. Nam cum dico Socrates et Plato vel ali-
quam individuam substantiam nomino, hoc aliquid significo.’
Boethius on General Terms 27
The species of the substantial form that is called ‘humanity’ is not con-
sidered as subordinate to the genus but as the same quality showing the
14 In Isag. II, 234, 14–235, 14: ‘At vero individuorum proprietas nulli communis est. Socratis
enim proprietas, si fuit calvus, simus, propenso alvo ceterisque corporis lineamentis aut
morum institutione aut forma vocis, non conveniebat in alterum’.
15 In De Int. II, 137, 6–16: ‘Incommunicabilis Platonis illa proprietas Platonitas appellatur. Eo
enim modo qualitatem hanc Platonitatem ficto vocabulo nuncupare possimus, quomodo
hominis qualitatem dicimus humanitatem. Haec ergo Platonitas solius unius est hominis.
16 In this book I will use the term ‘humanity’ in the sense it has in the following sentences:
‘man is a name of many things, humanity is a name of an attribute of those things’
(J. Stuart Mill, A System of logic ratiocinative and inductive, J.W. Parker, London 1863,
Vol.1, p. 34); ‘the definition of humanity will be corporeity and animal life, combined
with rationality, and with such and such a shape’(ibid., p. 185).
28 chapter 2
This short text contains an apparently incidental observation about the differ-
ence between ‘humanity’ and the quality called in the same way. The following
text however says otherwise, i.e. that humanity as species is a particular kind
of quality:
If the same humanity as species is a certain quality, why is it said that the
species is predicated essentially (it seeming a certain quality because of
some property of its nature)? We answer that the difference is only qual-
ity, humanity is not only quality, but is achieved through a quality.18
So, the difference that determines the genus is a quality; the species in a sense
is a quality, but, from another point of view, it presents itself as an additional
qualitative element, or, more precisely, as something ‘achieved through a qual-
ity’. This last sentence on the meaning of generic and specific terms is rather
enigmatic. For example, we might think that: 1) the qualitative meaning of
nominal expresions (NEs) is the most suitable candidate to play the role of
species-genus; 2) the same NEs, considered as proper nouns (i.e. in suppositio
simplex) refer to something different from the qualitative meaning they have
as common nouns (in suppositio personalis); 3) NEs have a meaning that takes
into account the meaning of specific and generic terms, and their categori-
cal specificity. I would say that in the texts we are analyzing point (1) prevails,
whilst in dealing with the problem posed by Porphyry the second aspect or
perhaps the third prevails.19 However, I do not believe that a line can be traced,
17 In Isag. II, 200, 7–16: ‘Alia est enim species substantialis formae quae humanitas nuncu-
patur, eaque non est quasi supposita animali, sed tamquam ipsa qualitas substantiam
monstrans […] substantialem determinat qualitatem.’.
18 In Isag. II, 327, 21–328, 9: ‘Si humanitas ipsa quae species est, qualitas quaedam est, cur
dicatur species in eo quod quid sit predicari, cum propter quandam suae naturae proprie-
tatem quaedam qualitas esse videtur? Huic respondemus quia differentia solum qualitas
est, humanitas vero non est solum qualitas’.
19 I say ‘maybe’ thinking of a possible reading of the texts: Boethius concludes his analysis of
Porphyry’s problem, arguing that species and genera are similarities. Apart from the diffi-
culty implied by this position, (3) might contribute to explaining how general terms refer
to objects that present similar characteristics, being categorially different. Terms such as
‘man’ and ‘white’, referring to species, could both mean similarity; the first the similarity
in being human the second in being white.
Boethius on General Terms 29
in the sense that Boethius limits himself to delineating the different perspec-
tives by discussing different arguments, at least from his point of view.
The data we have collected on the meaning of nominal expressions can be
(re) assembled as follows:
20 Boethius contrasts the classification in two grammatical categories with the Aristotelian
one in ten classes: ‘in quantum vox ipsa quidem intellectus significat, in duas (ut dictum
est) secatur partes, nomen et verbum. In quantum vero vox per intellectuum medieta-
tem subjectas intellectui res demonstrat, significantium vocum Aristoteles numerum in
decem praedicamenta partitus est’ (Ibid.).
30 chapter 2
depends on the fact that the determinative meaning of the predicate inhere in
the objects named by the subject.
In Chapter 9 of De Interpretatione Aristotle formulates his well-known defi-
nition of truth. The translation of Boethius is as follows:
The example proposed for an affirmative sentence is ‘dies est’ (‘the day is’). This
statement means by virtue of its terms that something is inherent in something
else, or, from another point of view, the meaning of ‘est’ applies (the Latin term
is ‘to add’) to the meaning of ‘dies’. In this case the verb has a full meaning and
refers to things that are, unlike when it is used in its copulative function. In the
negative proposition ‘dies not est’ the meaning of the predicate still applies to
that of the subject but the meaning of the verb ‘is subtracted’ from the mean-
ing of ‘day’. In the following text Boethius takes into account different types of
sentences:
21
In De Int. II, 204, 18–205, 30: ‘Si omnis adfirmatio vel negatio vera vel falsa est, et omne
necesse est vel esse vel non esse. […] Nam si verum est dicere quoniam album vel non
album est, necesse est esse album vel non album; et si est album vel non album, verum
est adfirmare vel negare.’
22
In De Int. II, 122, 7–11: ‘Adfirmatio namque in duobus terminis constans, aliquid alicui
inesse significat, tota autem vis ipsius est esse aliquid adnuere; negatio quoque aliquid
alicui non inesse significat, sed tota vis ipsius est abnuere atque disiungere’. The lectio I
have adopted (forced by my limits of understanding) is middle ground between the criti-
cal edition and the text of Patrology: In De Int. 455 C.
Boethius on General Terms 31
23 In De Int. II, 140, 29–141, 11. Cf. also 188, 19–21: ‘Plato philosophus est, haec enim propositio
Platoni philosophiam inesse constituit […] cum dicimus homo albus est, albedinem uni-
versali rei inesse monstramus, id est homini, aut certe particulare ac singulare, ut si quis
dicat Socrates albus est, albedinem enim Socrati singulari substantiae et proprietati in-
communicabili inesse significavit.’ The last sentence suggests a possible intensional read-
ing of the predication: Boethius says that the whiteness inheres in the singular substance
and in an unshared property of Socrates.
24 I would distinguish the considerations that Boethius makes in the De Interpretatione pas-
sages from those closely related to Aristotle’s text of In Cat 5: ‘… ut album de corpore
predicatur, dicitur enim corpus album’. In the former case, Boethius considers a relation
between a property and the individual having that property, in the latter case a relation
between non-interpreted terms – a relation that can be rendered in an abstract way by
the formula /album/ R /corpus/.
32 chapter 2
So far we have arranged the text material to discover a path leading to a formu-
lation of the P1 principle. But what about P2? We would need texts in which
an analysis of universal sentences comes alongside that of the relationship of
predication. What I have been able to discover is an analysis of the ambiguity
of the determiner omnis. The reader can find some information on this issue
below, but I think the material I have ‘extracted’ does not make it possible to
clarify precisely the logical-semantic role of quantification or to link a quanti-
fied sentence to a formula that establishes a relationship of predication be-
tween the terms of the same sentence.
In his commentary on De Interpreatione, Boethius considers different
types of grammatical subject, in particular common nouns with a determiner
(signum quantitatis) and proper nouns. ‘Homo’ is considered an ambiguous
term as it can refer to the abstract entity (whose name is ‘(the) man’) or to
some particular man, as ‘nomen hominis et de omnibus dici potest et de sin-
gulis quibusque qui sub una humanitatis specie continentur’.25 The ambiguity
is resolved by putting a determiner like ‘omnis’ or ‘quidam’ before the noun.
Syncategorematic expressions determine the ‘distribution’ of the universal
term in a universal or particular way.
Speaking of univocity of affirmations and negations (commenting on De
Int. 8), Boethius presents the question: ‘quemadmodum universalis affirmatio
unam rem significare possit cum ipsa universalitas non de uno sed de pluribus
praedicetur’.26 Since ‘every man is white’ means a multiplicity of men, it would
appear that several things are meant in the same sentence. Boethius’s labori-
ous response is that ‘universal’ is said of the sentence as it means the unity
of being man and not individual men.27 In this case, some kind of formal or
collective unity is at stake; at the same time Boethius considers a distributive
25
In De Int. II, 138, 1–3.
26
In De Int. II, 178, 27–179, 2.
27
In De Int. II, 179, 12–15: ‘cum dicitur omnis homo justus est, non tunc singulos intellegi-
mus, sed ad unam humanitatem quidquid de homine dictum est dicitur.’
Boethius on General Terms 33
Boethius writes two commentaries on Porphyry. In the first one he uses the
translation of the Isagoge by Marius Victorinus, in the second one he trans-
lates Porphyry’s text directly and provides an extensive commentary. Boethius
tries to answer one of the question about predication (see 1.4 above), concern-
ing the ontological status of the things that are said-of. In this section we will
focus on the crucial passages of Boethius’s two commentaries concerning the
solution of Porphyry’s question. Let me begin with a preliminary clarification
of the meaning of the terms ‘species’ and ‘genus’. Substances and accidents
are called species and genera in that they have a relationship with something
other:
The relation with something other is constitutive of the genus, just like
/animal/, if it is asked about its substance, one has to say /animal/ is
/sentient animate substance/, so that this definition shows the thing as it
is in itself, not as referring to something other; when, on the other hand,
one says that /animal/ is a genus, I do not think the thing is being referred
to, but the relation by which /animal/ can be linked to other things which
are subordinate to it. The nature and form of the genus consists in the
fact that it is in relation by predication with those things in the substance
of which is predicated, they being many and different by species. 29
28 For instance, when, commenting Porphyry’s text, Boethius says that /substance/ is predi-
cated among other things of /corporeal/, as ‘what is corporeal is said substance’. However,
also in this case, the relation of predication is not clearly interpreted.
29 In Isag. II, 182, 16–183, 4: ‘Ad alterum relatio genus facit, ut ipsum animal, si eius sub
stantiam quaeras, dicam substantiam esse animatam atque sensibilem. Haec igitur defi-
nitio rem monstrat per se sicut est, non tamquam referatur ad aliud. At uero cum dicimus
animal genus esse, non, ut arbitror, tunc de re ipsa hoc dicimus sed de ea relatione qua
potest animal ad caeterorum quae sibi subiecta sunt praedicationem referri. Itaque char-
acter est quidam ac forma generis in eo quod referri praedicatione ad eas res potest, quae
cum sint plures et specie differentes, in earum tamen substantia praedicatur’. Among
34 chapter 2
‘Subordinate’ here has a metaphorical meaning and it does not say much as
to the relation of predication. The philosophical definition is the one referred
to at the end of the quotation: ‘the genus is what is predicated in essence of
several things differing in species’.31
the numerous texts available, I limit myself to quoting this one because it is particularly
explicit.
30 Boethius’ transl., In Isag. II, 178, 14–178, 18: ‘genus dicitur cui supponitur species, ad horum
fortasse similitudinem dictum. etenim principium quoddam est huiusmodi genus earum
quae sub ipso sunt specierum, videtur etiam multitudinem continere omnem quae sub
eo est’; 179, 18–180, 3: ‘tripliciter igitur cum genus dicatur, de tertio apud philosophos
sermo est; quod etiam describentes assignauerunt genus esse dicentes quod de pluribus
et differentibus specie in eo quod quid sit praedicatur’.
31 Boethius associates the notion of universals to the Aristotelian one of ‘thing that is said
of’: ‘Porphyry divides the things that are predicated … into those that are predicated of in-
dividual things and those predicated of many things … as for instance Socrates, Plato, this
white in this snow, this chair I am sitting on, not the universal chair, … <the> white and
<the> snow are universals, not this white in this snow; the latter cannot be predicated of
Boethius on General Terms 35
If we observe the path and the order, once the genus of anything is found,
another genus is required, and yet another one […]. It is necessary that in
the ascending chain we reach a certain something that, being the genus
of other things, has no other genus over itself … but if that happens for
the genus, no doubt in the descent through the species we will rest at the
end, having met a final point.33
The connection with the lower things is clarified by the second characteriza-
tion of the species as what is predicated in eo quod quid sit of several numeri-
cally different things. The relation ℘ can be constructed by answering the
question ‘what is x’, where the values of ‘x’ are individual things (e.g. ‘man is
predicated of Cicero and Demosthenes and other numerically different things
as all of them are men).34 No particular restriction is set on the typology of
individual things that can fall under a certain species. As to the number of
any other white thing … there are things that can be predicated of several things, e.g. the
species, genera, etc.’ (In Isag., 183, 15–184, 13) [Omnium quae praedicantur quolibet modo,
facit Porphyrius diuisionem idcirco, […] alia de singularitate, alia de pluralitate dicuntur.
[…] ut Socrates, Plato, ut hoc album quod in hac proposita niue est, ut hoc scamnum in
quo nunc sedemus, non [omne] scamnum – hoc enim uniuersale est […] uniuersale est
enim album et nix – sed hoc album quod in hac niue nunc esse conspicitur; hoc enim non
potest de quolibet alio albo praedicari […] Alia uero sunt quae de pluribus praedicantur,
ut genera, species …].
32 Relation also called ‘connexion’ and ‘comparison’: In Isag. 65, 4–5.
33 In Isag. I, 66, 18–67, 5: ‘Sed si in uno filo atque ordine speculemur et quodcumque
genus alicuius rei repertum sit, eius rursus genus aliud requiramus et rursus aliud […].
Necessario est posse nos ascendentes usque ad tale aliquid peruenire cuius, cum ipsum
caeteris genus sit, aliud genus inuenire non possumus […] Sed si hocin genere contingit,
ut ascendentes alicubi consistamus, non est dubium quin descendentes iterum per spe-
cies ad aliquem quodammodo calcem offenso termino consistamus’.
34 In Isag. 205, 4–7.
36 chapter 2
substantia
corporea incorporea
corpus
animatum inanimatum
corpus animatum
sensibile insensibile
animal
rationale irrationale
animal rationale
mortale immortale
homo
Looking at the figure one cannot help thinking that a good selection of the
subaltern species depends on a restricted interpretation of the predication in
eo quod quid sit. The identification of a specific node clearly has to do with the
35 An element x in an order relation R is said to be minimal if and only if there is no other
element in R which precedes x. Conversely, Aristotelian categories (substance, quality,
quantity, etc.) can be considered as maximal elements, given the definition: an element x
in an order relation R is maximal if and only if x follows every other element in R.
36 The examples Boethius supplies are not without problems: ‘sun’, ‘world’, ‘moon’ are ob-
jects related to different species, ‘Phoenix’ corresponds to a species whose only object
belongs to mythology: Cf. In Isag. 218, 6–7.
Boethius on General Terms 37
37
In Isag. 208, 9–17: ‘His igitur cognitis sumamus praedicamenti unius exemplum. Ut ab
eo in caeteris quoque praedicamentis atque in a caeteris speciebus in uno filo atque or-
dine quid eueniat possit agnosci. Substantia igitur generalissimum genus est; haec enim
de cunctis aliis praedicatur. Ac primum huius species duae, corporeum, incorporeum;
nam et quod corporeum est, substantia dicitur et item quod incorporeum est, substantia
praedicatur. Sub corporeo uero animatum atque inanimatum corpus ponitur, sub ani-
mato corpore animal ponitur’.
38 chapter 2
I leave aside the question of whether the genera and species subsist or are
only placed in bare concepts, or whether they are subsistent corporeal or
incorporeal things, and whether they are connected with the objects of
perception or separate from it.38
The intellect conceives the species /man/ under /animal/ imagining that
it includes individual men, and that incorporeal thing of which it initially
considered the corporeal particulars in the individual men (that are) ob-
ject of senses and understanding. In fact that special man including us
all within the domain of the name signifying it, is not to be thought of
as something corporeal, since it is understood by the mind only and the
intellectual faculty.39
38
In Isag. I ed.159, 3–7: ‘de generibus ac speciebus illud quidem sive subsistent sive in solis
nudisque intellectibus posita sunt sive subsistentia corporalia sunt an incorporalia et
utrum separata a sensibilibus an in sensibilibus posita et circa ea constantia, dicere re-
cusabos’. ‘Sensibilia’ translates here and afterwards as ‘objects of perception’; sometimes
I shall use ‘sensible (perceptible) things’ for stylistic reasons. The intended meaning is:
‘the entities we attend to when we perceive the world … using our five senses … The ob-
jects of perception include such familiar items as paper clips, suns and olive oil tins. It
is these things themselves that we see, smell, touch, taste and listen to’ (Daniel O’Brien,
‘Objects of Perception’, in: Internet Encyclopedia of Philosophy: http://www.iep.utm.edu/
perc-obj/).
39
In Isag. 24, 20–25, 3: ‘(intelligentia) speciem ipsam hominis, quae sub animali est posi-
ta, et singulos homines continere suspicatur et illud incorporeum intellegit cuius ante
Boethius on General Terms 39
Such a question is solved in a non trivial way along these lines: […] /incor-
poreal/ can be in some way related to /corporeal/ […] since /substance/
is a genus, /corporeal/ and /incorporeal/ are species of /substance/;
/<in>corporeal/ is not a genus <of /substance/>, because /incorporeal/
falls under /substance/ genus. […] For this reason </substance/> genus
is such that it is neither corporeal nor incorporeal […] some species
are corporeal, some other incorporeal; in fact, if you place /man/ under
/substance/, you have placed a corporeal species, if (you place) /God/ an
incorporeal one. 42
40 In Isag. 26, 13–20: ‘an corporalia ista sint an incorporalia. […] incorporalia esse quae ipsa
quidem nullis sensibus capiantur, animi tamen qualia sint consideratione clarescunt.
Nam quia incorporeorum prima natura est, potest res incorporea parens esse quodam-
modo corporeae. Corporea uero incorporeis praeesse non poterunt’.
41 For example, if the man is considered as a node of the predication relation, what is said of
the man – (for instance being the worst inhabitant of the world or being a species) does
not necessarily apply to the individual under it (like Socrates might be).
42 In Isag. 26, 15–27, 20: ‘Haec quoque non mediocriter utilis inquisitio ita resoluitur: […]
potest res incorporea parens esse quodammodo corporeae. […] quoniam substantia
genus est, corporale uero et incorporale species substantiae, corporale non esse genus
haec res declarat, quod substantiae, id est generi, incorporale supponitur […] ergo genus
Boethius on General Terms 41
However, one may doubt that Porphyry had this in mind. Boethius seems to be
acquainted with this suspicion:
But those to whom the same integral whole seems to be at stake, consider
such things (=universals) only incorporeal. They do not regard a genus as
a certain thing having its own nature, but as a genus. So, if /substance/ is
a genus, it is not considered as a substance, but as having some species
under itself. 43
This text is not easy to interpret: a thing (for instance /man/) should be con-
sidered differently according to a certain property or point of view. Boethius is
referring to sentences like:
/man/ is a species
/animal/ is a genus (i.e. has some species under itself)
/species/ (/genus/) is an incorporeal entity
tale est, ut ipsum neque corporale neque incorporale sit, […]Species alias corporalis, alias
incorporalis est. Nam si hominem sub substantia ponas, corporalem speciem posuisti, sin
deum, incorporalem’.
43
In Isag. 28, 14–29, 3: ‘Sed sunt quibus hoc ipsum integrum uideri possit, et haec solum
incorporalia esse definiunt. Qui sic dicunt, non considerari genus in eo quod quaeque res
suapte natura constat sed in eo quod genus sit. Itaque si substantia genus est, non consi-
deratur in eo quod substantia est sed in eo quod sub se species habet’.
42 chapter 2
not follow that a∈c.44 Assuming this reading, we get the result that in some
‘external’ branch of the predication tree /man/ falls under /incorporeal/ but /
incorporeal/ is not predicated of it, and in some other ‘internal’ branch /man/
falls under /corporeal/ and /corporeal/ is predicated of it. The solution, in so
far as it works,45 would have the advantage of justifying the exclusivity of the
alternative given by Porphyry and at the same time distinguishing between
abstract and concrete universals.
The ‘sed sunt quibus’ leads one to believe that it is a matter of other people’s
opinion; this could match the fact that Boethius has just suggested considering
the alternative mentioned as non-exclusive. Besides, when Boethius comes to
consider the next question, he notes how it makes sense only if one accepts
the idea that genera and species are incorporeal things (and then, implicitly,
that Porphyry speaks of ‘only integral wholes’). In presenting this side of the
problem, Boethius makes an interesting use of the term ‘thing’:
Some incorporeals are things without a body, as the soul or God; others
cannot be without a body, as the first incorporeity after the terms (ex-
tremities of geometrical figures); others are in the bodies and have their
own being regardless of bodies, such as the soul. 46
One may wonder what category of incorporeal things universals belong to.
Boethius provides this answer:
It seems that (genera and species) can sometimes be separate, other times
not. When one makes the division of genera and species of corporeal
44 P.V. Spade, (‘Boethius against Universals: The Arguments in the Second Commentary on
Porphyry’ p. 12–13. web: pvspade.com/Logic/docs/boethius.pdf) believes Boethius may
consider a predication relationship between entities of different logical levels in mind.
Even putting things this way, transitivity is lost.
45 One can observe that ‘species’ and ‘genus’ are relative terms and denote (ordered pairs of)
things such as </man/, /animal/> and so on. The predicate ‘is a species’ can be handled
as a one-place predicate, but it is not so easy: taking advantage of the language of the
predicate calculus enriched by lambda terms, ‘is a species’ could be rendered with �λx∃y
x<y where < expresses the species-of relation. Otherwise, one can simply forget about the
categorial order induced by the predication relation, something that is easy for us (we
actually speak of natural kind context free), but probably not so easy for Boethius.
46 In Isag. 29, 22–30, 4: ‘quaedam incorporales sunt res, quae omnino corpora non patiuntur,
ut animus uel deus, quaedam uero quae sine corporibus esse non possunt, ut prima post
terminos incorporalitas, quaedam autem quae in corporibus sunt et praeter corpora sese
esse patiuntur, ut anima’.
Boethius on General Terms 43
things and mentions their propers and differences, no doubt genera and
species are about objects of perception, i.e. corporeal; but when incorpo-
real things are concerned and one makes a division among them, (genera
and species) incorporeal things are at stake. So, there is no doubt these
five things belong to the same genus, since they can be apart from bod-
ies and connected with bodies, in such a way that if they are connected
with bodies, they are inseparable from them, if connected to incorporeal
things they are not separable from incorporeal things.47
What does ‘be about’ mean exactly? In addition, in what sense does Boethius
distinguish between the species and genera of corporeal and incorporeal
things? I would be inclined to read ‘to be with reference to’ as ‘to be in the
relation ℘′. The question about the separation or immanence of the species/
genus is subsequent to that concerning their immateriality and presupposes
an affirmative answer to the latter. Thus, I believe Boethius is speaking of the
form (property) of being human, instead of the generic man, of which form ‘to
be abstract’, ‘separate’, ‘immanent’ can be soundly predicated.
The form of man, considered as an incorporeal entity, can be located in in-
dividual men; the form of soul in living beings (or in the individual souls). The
next statement about the fact that the five Porphyrian expressions belong to
the same genus leads us back to the question of what the predication tree must
be: the common genus could be /incorporeal/, the species of which are /sepa-
rable from bodies/ and /inseparable/. The latter forms cannot be nodes of the
traditional Porphyrian tree, but presuppose a certain ‘outside’ ramification. Let
us consider the predication principles:
47
In Isag. 30, 7–17: ‘Videtur autem quod et segregari et iungi possint. Nam quando corpora-
lium diuisio per genera in species fit et eorum propria et differentiae nominantur, haec
circa sensibilia, id est corporalia esse non dubium est; cum uero de incorporalibus rebus
tractatus habetur et per ea ipsa diuiduntur quae corpore carent, circa incorporalia uer-
santur. Quodsi hoc est, non est dubium quod quinque haec ex eodem sunt genere, quod
et praeter corpora separata esse possint et corporibus iungi patiantur sed ita, ut si cor-
poribus iuncta fuerint, inseparabilia a corporibus sint, si uero incorporalibus, numquam
ab incorporalibus separentur’.
44 chapter 2
Form
of animal
Form
of man
Only the red and black arrows are – in this order – composable. This fact guar-
antees, on the one hand, that Socrates is an animal, Socrates being a man and
the form of man subordinated to that of animal; on the other, it avoids the
predication of ‘incorporeal’ with respect to individual men.
At this point, we may wonder again about the instances of predication prin-
ciple. Boethius says that ‘genera and species are about particular things’ and
‘they are separable from bodies’ (quinque … praeter corpora separata esse pos-
sint); One can infer that some relation of being-in has to exist between indi-
vidual things and universal entities. The problem is that we do not find any
in-depth analysis of this issue, so it is difficult to reason about the concept of
inherence or any instance of (individual) predication principle in the style of
twelfth-century realist philosophers.
Boethius on General Terms 45
48 On different senses of ‘common’ P.V. Spade, (op. cit., p. 4) recalls Porphyry’s Exposition of
Aristotle’s Categories by Question and Answer (Greek text and translation in Appendix 2).
46 chapter 2
Therefore, none of the ordinary meanings of ‘common’ is the right one. Genera
and species are said to be common because they ‘are totally’ in several things
at the same time as constitutive elements of their essence. Premise (2) of the
argument invokes a metaphysical theory of substance: a certain thing is not a
genus or species only (or mainly) because it is predicated of something other,
but also (or mainly) because it is a constitutive element of the thing of which
it is predicated.
In order to define the relation of metaphysical foundation some decision on
what we consider in some sense real seems to be required. If existing things are
minimal elements of the predication relation, the intermediate elements – to
be such – must be constitutive of the first ones, without counting as adjunctive
elements of reality. For instance first substances are counted as elements of
reality, they are count unities; common things are not elements that count as
additional items in the ontological basket.
Boethius provides a second argument against the existence of genera and
species: while the first argument starts from the thesis that universals are
not unitary and from the postulate of existence, the second assumes both
the possible alternatives (genera are unitary – multiple) and, by negating the
consequent, concludes that genera and species do not exist. Boethius’ line of
reasoning can be schematized as follows:
49
In Isag. 162, 15–24: ‘Una enim res si communis est aut partibus communis est et non iam
tota communis sed partes eius propriae singulorum; aut in usus habentium etiam per
tempora transit ut sit commune ut seruus communis uel equus; aut uno tempore omni-
bus commune fit, non tamen ut eorum quibus commune est substantiam constituat, ut
est theatrum uel spectaculum aliquod, quod spectantibus omnibus commune est. Genus
uero secundum nullum horum modum commune esse speciebus potest’.
Boethius on General Terms 47
Assuming genera and species are multiple, we are faced with the following
difficulties:
As numerous as species may be, the genus in them is one, not in such a
way that they seize parts of the genus, but individual species have the
whole genus at one time; it follows that genus cannot be a single thing,
being in several individual things at the same time.50
Porphyry states: either species-genera are in the concepts only or they are
subsisting entities. Boethius shifts the attention to the relationship between
50 In Isag. 161, 20–24: ‘Quantaecumque enim sunt species in omnibus genus unum est, non
quod de eo singulae species quasi partes aliquas carpant sed singulae uno tempore totum
genus habent. Quo fit ut totum genus in pluribus singulis uno tempore positum unum
esse non possit’.
51 I agree then with P. King’s conclusion, according to which Boethius’ discussion presents
itself as ‘a dialectical investigation based on hypotheses in the style of the second half
of the Parmenide’ (‘Boethius’ anti-realist arguments’, B. Inwood (ed.), Oxford Studies in
Ancient Philosophy, vol XL, OUP, Oxford 2011, 381–401, p. 399). I am not so sure about the
distinction that King makes between two arguments (after the first criticism of realism):
looking at the schema suggested above one can see one or two arguments depending on
the path he chooses to follow starting from the first or the second vertical lines of the
schema.
52 In Isag. 163, 7–14: ‘omnis intellectus aut ex re fiat subiecta ut sese res habet aut ut sese
res non habet (nam ex nullo subiecto fieri intellectus non potest). Si generis et speciei
caeterorumque intellectus ex re subiecta ueniat ita ut sese res ipsa habet quae intellegi-
tur, iam non tantum in intellectu posita sunt sed in rerum etiam ueritate consistunt, et
rursus quaerendum est quae sit eorum natura’.
Boethius on General Terms 49
concepts and things in the world: the species (-genus) concept corresponds
either to one thing as it is or as it is not. Apparently there are two alternatives
again; Porphyry’s empty concept corresponds to the concept that does not rep-
resent things as they are in reality. At this point, however, Boethius – following
Alexander’s commentary – proposes to distinguish between sound and false
concepts:
We say that it is not necessary to consider false and empty every concept
deriving from some object, but not in such way as to represent it as it is in
reality; only in those concepts deriving from a composition, <can> a false
belief and not a true intellectual understanding occur.53
Genera, species and remaining objects are located in both corporeal and
incorporeal things […] the soul manages to separate the incorporeal na-
ture from the corporeal things and observes the pure form as it is in itself
[…]. Nobody then should say we conceive the line in an incorrect way,
when we conceive it as separated from bodies, whereas it is not actually
separated from the body; in fact not every concept that does not repre-
sent things as they actually are in reality is to be considered false.54
53
In Isag. 164, 5–8: ‘Non enim necesse esse dicimus omnem intellectum qui ex subiecto qui-
dem fit, non tamen ut sese ipsum subiectum habet, falsum et uacuum uideri. In his enim
solis falsa opinio ac non potius intellegentia est quae per compositionem fiunt’
54
In Isag. 165, 3–20: ‘Genera ergo et species ceteraque vel in incorporeis rebus vel in his
quae sunt corporea reperiuntur; […] (anima) aufert ut solet a corporibus incorporeorum
naturam et solam puramque ut in se ipsa forma est contuetur […] nemo ergo dicat falso
nos lineam cogitare, quondam ita eam mente capimus quasi praeter corpora sit, cum
50 chapter 2
Having reached this point, one may wonder whether any progress has been
made by moving from things to concepts. Certainly we have a way to establish
a predication principle for concepts and distinguish between concepts hav-
ing some form of correspondence with reality, and fictitious images obtained
by composition, but does this way of putting things provide us with any clue
to answering Porphyry’s question? In other words, was Porphyry interested in
the concepts or in the things the concepts refer to? Once we have established
that genera and species are in the bodies and are conceived apart from them,
the problem of what we conceive when we conceive something abstractly, still
remains .55
Boethius does not hide behind the conceptualist solution, and tries to
say something about the sort of entities the concepts of genera and species
refer to:
When one thinks of genera and species, one obtains some similarities of
the individual things in which they are; for instance, from different men
one gets the similarity of humanity, which becomes (fit) species once it is
thought of and represented in a true way.56
praeter corpora esse non possit, non enim omnis qui ex subiectis rebus capitur intellectus
aliter quam sese ipsae res habent, falsus esse putandus est’.
55 The archetypal characterization of thought as thought of something is in Parmenide, 132b,
c9: ‘thought cannot be thought of nothing’.
56 In Isag. 166, 8–12: ‘cum genera et species cogitantur, tunc ex singulis in quibus sunt eorum
similitudo colligitur ut ex singulis hominibus inter se dissimilibus humanitatis similitudo,
quae similitudo cogitata animo ueraciterque perspecta fit species’.
57 The mandatory reference is to the Neoplatonic theory of the triple way of being of the
universal, which dates back at least to Albino’s Epitomé or Didascalicon (cf. A.C. Lloyd,
‘Neoplatonic logic and aristotelian logic’, Phronesis 1955–1956, 59–62). Some interpreters
have drawn attention to the fact that Boethius does not take this theory into account
(cf., e.g. A. de Libera, L’art des généralités, Aubier, Paris, 1999, p. 200; J. Marenbon, Boethius,
Boethius on General Terms 51
Two things can actually be in the same object for different reasons, as the
concave and convex line,60 things that, being defined in a different way
and giving rise to different concepts, nonetheless are always found in the
same subject: in fact the concave and convex line are actually the same
line. The same applies to genera and species, that is to singularity and
universality. 61
Oxford U.P., Oxford 2003, 25). Other scholars give more relevance to the theory, for exam-
ple R. Cross says that Boethius never states the theory but it nevertheless ‘underlies and
gives structure to what Boethius has to say on the topic’ (‘Form and universal in Boethius’,
British Journal for the History of Philosophy, 20, 2012, 439–58). It must be remembered
in any case that Ammonius’s School (where the theory at issue was taught) represents
a primary source for Boethius (Cf. P. Courcelle ‘Boèce et l’école d’Alexandre’, Mélanges
dìarcheologie et d’histoire de l’Ecole francaise de Rome, 1935, 185–223; Les lettres grecques
en Occident. de Macrobe à Cassiodore, Paris 1948), hence the theory, if not endorsed by
Boethius, remains in his cultural background.
58 In Isag. 166, 19–21: ‘haec similitudo cum in singularibus est fit sensibilis; cum in uniuersali-
bus fit intellegibilis – eodemque modo cum sensibilis est in singularibus permanet; cum
intellegitur fit uniuersalis’.
59 Supposing that the thing is the same. In medieval times a criticism like this will be ad-
dressed to the similitudo theory, which derived from an interpretation of the Boethian
doctrine.
60 The example is Aristotelian (De an. III, 431 b12–16); see A. De Libera, op. cit., 30; 218.
61 In Isag. 166, 23–167, 4: ‘Neque enim interclusum est ut duae res eodem in subiecto sint
ratione diuersae, ut linea curua atque caua, quae res cum diuersis definitionibus termi-
nentur diuersusque earum intellectus sit, semper tamen in eodem subiecto reperiuntur;
52 chapter 2
eadem enim linea caua, eadem curua est. Ita quoque generibus et speciebus, id est singu-
laritati et uniuersalitati’.
62 I do not know if this correspondence must be maintained for similarity meant as a real
property. In the context of the present discussion it is possibly not the case that common
properties of substances, being categorially different from what they are properties of,
involve an unwanted multiplication of entities. It is a particularly complicated issue of
Aristotelian exegesis: accidental properties are-in the first substances; besides there are
qualities which are-in and said-of (i.e. universal qualities); and finally essential qualities
that are-not-in substances, as being-in characterizes the accident.
Boethius on General Terms 53
63
In Isag. 167, 7–12: ‘His igitur terminatis omnis, ut arbitror, quaestio dissoluta est. Ipsa enim
genera et species subsistunt quidem alio modo, intelleguntur uero alio. Et sunt incorpora-
lia sed sensibilibus iuncta subsistunt in sensibilibus. Intelleguntur uero ut per semet ipsa
subsistentia ac non in aliis esse suum habentia’.
54 chapter 2
‘To be continued’, in some sense.65 Despite this, Boethius may believe he has
made some progress from forms to similarities: the former present themselves
at the same time as individual and as common objects (according to a certain
interpretation of the Aristotelian text); the latter too can be said to be com-
mon, in the apparently less problematic sense in which one says that a relation
is about several things. However if we interpret the similarities as forms, we
are compelled to provide many philosophical justifications. Universals that are
apparently less problematic are those introduced and immediately put aside
in the first commentary: the generic man, the generic animal; it simply does
not make sense to say the generic man is in the individual man, for any sense
of ‘being-in’.
Boethius no longer intends to deal with things that are not directly related
to Porphyry’s query, at least in the present context. On the one hand, he has
explained the properties of the predication relation at an abstract level; on the
other hand, he has supplied various ontological options about what can in-
stantiate the terms of the relation. We have discovered what is said-of: the sim-
ilarities, entities that do not match any entry in the standard categorial system.
In Boethius’ commentaries on Porphyry there are no explicit instantiations of
the Aristotelian principles P1-P2, but an answer, or perhaps several answers, to
one of the possible questions related to the issue of predication.
64 In Isag. 167, 12–17: ‘Sed Plato genera et species caeteraque non modo intellegi uniuersa-
lia, uerum etiam esse atque praeter corpora subsistere putat, Aristoteles uero intellegi
quidem incorporalia atque uniuersalia sed subsistere in sensibilibus putat. Quorum dii-
udicare sententias aptum esse non duxi, altioris enim est philosophiae’.
65 One possibility is to proceed by taking into account the texts where Boethius distinguish-
es several types of entities that are the subject matter of specific disciplinary analysis, and
different modalities of being (De Trinitate II, De Institutione arithmetica I, 1, and, coming
back to Porphyry’s comments, In Isag. I, 8–9), texts that are well known and largely quot-
ed in historiography. The archetypal distinction between disciplinary objects is in Met. VI,
1 1026a 10–16. On the degrees of speculative knowledge and on the Boethian doctrine of
being, see L. Obertello, Severino Boethius, Accademia ligure di scienze e lettere. Genova
1974, parte IV, cap. I, III.
chapter 3
3 Incidentally, even Heidegger has his place in the vast literature on Scotus: cf. P.W. Rosemann,
‘Causality as Concealing Revelation in Eriugena: a Heideggerian Interpretation,’ American
Catholic Philosophical Quarterly, 79 (2005), 653–671. In regard to whether or not Eriugena
is a systematic thinker, J. Marenbon notes that ‘the thought of the Periphyseon does form
a system, but one which could be called ‘philosophical’ only in an unhelpfully broad sense
of the word. It is a system which does not attempt to provide an explanation of reality by
means of reason, but rather, to make an imaginative whole of ideas, arguments and dogma
taken from a variety of sources, including Holy Scripture’: ‘John Scotus and the Categoriae
decem’, in Eriugena: Studien zu seinen Quellen. Edited by Beierwaltes Werner. Heidelberg: Carl
Winter 1980; 116–134 (repr. in: J. Marenbon, Aristotelian Logic, Platonism and the Context of
Early Medieval Philosophy in the West, Aldershot, Ashgate, 2000); p. 133.
4 My intent is in some respect similar to C. Erismann’s in his doctoral thesis (La Genese réal-
isme ontologique durant le haut du Moyen Age, Lausanne 2006) but more modest than the one
by the same author in his article ‘Alain de Lille Metaphysique the érigénienne the pluralité
et des formes’ (in J. Solère – A. Vasiliu – Galonnier A. (eds), Alain de Lille, le Docteur univer-
sel. Philosophie, théologie et littérature au XIIe siècle. Actes du XIe Colloque international de
la Société internationale pour l’Étude de la Philosophie médiévale, Paris, 23–25 octobre 2003,
Brepols, Turnhout 2005. pp. 19–46). In this second work Erismann tries to connect Eriugena,
Alain of Lille and Thierry of Chartres. The first link is textually documented, but the use
Erismann makes of it is rather unnatural: according to him Scotus, as a realist, influences
an apparently nominalist theory. The second link is more natural but with little documenta-
tion (Scotus’ influence on the school of Chartres has, in fact, been challenged by E. Jeauneau
in ‘Le renouveau érigénien du XIIe siècle,’ in W. Beierwaltes (ed.), Eriugena redivivus. Zur
Wirkungsgeschichte seines Denkens in Mittelalter und im Übergang zur Neuzeit (Vorträge des
V. Internationalen Eriugena-Colloquiums Werner-Reiners-Stiftung Bad Homburg 26–30. August
1985), Carl Winter Universitätsverlag, Heidelberg 1987, 26–46; see pp. 44–45.
5 There are numerous studies on the tradition of the Periphyseon which show that the text
(or parts of it) was relatively widely circulated between the eleventh and twelfth centuries.
See inter alia: J. Marenbon, From the Circle of Alcuin to the School of Auxerre, Cambridge UP,
Cambridge 1981; D. Moran, The Philosophy of John Scottus Eriugena 267–277; E. Jeauneau, Le
renoveau quot.
The Metaphysical System of Scotus Eriugena 57
3.2 Nature
6 This ‘absolute’ meaning is suggested by Boethius in the Liber contra Eutychen et Nestorium.
See, on this subject, D. O’Meara, ‘The Concept of Natura in John Scottus Eriugena (Periphyseon
naturae Book I),’ Vivarium XIX, 2 (1981), 126–145, in particular pp. 128–29 (in the latter pages
the author also draws attention to the role of Pseudo-Dionysius in the use of ‘nature’ for thing
which are and are not); M. Cristiani, ‘Nature-essence et nature-language. Notes sur l’emploi
du terme ‘natura’ dans le Periphyseon de Jean Scot Erigène.’ In J. Beckmann (ed.), Sprache
und Erkenntnis im Mittelalter. Akten des VI. internationalen Kongresses für mittelalterliche
Philosophie der Société internationale pour l’étude de la philosophie médiévale, 29. August–3.
September 1977 im Bonn. Walter de Gruyter, Berlin 1981. pp. 707–717. Cr. in particolar p. 711).
7 Periphys., I, 3, 19–4, 22: ‘Videtur mihi divisio naturae per quattuor differentias quattuor spe-
cies recipere, quarum prima est in eam quae creat et non creatur, secunda in eam quae et
creatur et creat, tertia in eam quae creatur et non creat, quarta quae nec creat nec creatur’.
8 I.P. Sheldon-Williams suggests a Pythagorean numerical theory as a source for this divi-
sion into four species. This possible origin is referenced in Philo, quoting a passage from
Philolaus, who is utilizing the De mensibus of Lydus: cf. ‘Johannes Scottus Eriugena’, Chap 34
of A.H. Armstrong (ed.) The Cambridge History of Later Greek and Early Medieval Philosophy,
p. 522. S. Gersh, on the other hand, believes that the sources of this Pythagorean theory are
to be found in Macrobius’ Commentarius in Somnium Scipionis and in De Nuptiis Philologiae
et Mercurii by Martianus Capella (‘Eriugena’s fourfold contemplation. Idealism and
arithmetic,’ in S. Gersh – D. Moran (eds.), Eriugena, Berkeley and the Idealist Tradition,
University of Notre Dame Press, Notre Dame 2006, 151–167). The latter author believes that
the numerical theory underlying the distinction between the four natures is the key to read-
ing all of Scotus’ work. In this connection he presents a series of quotations occurring at
different places in Periphyseon.
58 chapter 3
that there is nothing more reasonable than dealing with such a distinction im-
mediately, ‘not only because it is the first one to be met, but also because it
appears to be and is supremely obscure.’9
In a first sense, those things are said to be which are perceived by the senses
or by the intellect (perceptio intelligentiae).10 Among these things are bodily
accidents. Those which are not ‘because of the excellence of their nature’ can-
not be objects of perception, intellect or reason. God and created essences are
things of this kind. Scotus is fond of paradoxes, and in this case he attaches a
kind of metaphysical excellence to things that are not. Moreover, he states that
what is not is completely unknowable, even though he does have several things
to say about it.11
The second sense in which things are said to be or not to be has to do with
a hierarchy of entities ordered according to their level of perfection. The first
element is the most sublime of all angels; the last one is the ‘extreme part of
the rational and irrational soul which is said to be nourishing or active life.’12
Scotus does not clarify what properties are shared by things or classes of things
(ordines) in the hierarchy, nor how the hierarchy itself is to be understood,
apart from the insight on the greater or lesser sublimity of ordered things (and
classes of things).
According to Scotus, affirming something of a certain subject is inferior
to denying the same thing: if man is a visible, mortal, rational animal, angels
(assuming they are superior) are not visible, mortal, rational animals;
9 Periphys., I, 5, 48–50.
10 Scotus’ psychology is rather complicated. The traditional distinction between sense,
imagination, intellect and reason is maintained; additionally, however, there are process-
es that allow for the manipulation of specific objects of a faculty or function of the soul
by another faculty or function. In this particular case it seems that the objects of sense
faculties are ‘seen’ by the intellect while maintaining their status as objects of the senses.
It is most likely for this reason that Scotus speaks of intellectual perception.
11 Scotus clearly draws from the tradition of negative theology, in particular that of
Dionysius the Areopagite. Among the numerous explorations of the subject, see Moran,
The Philosophy., quot., p. 212–240 (on the tradition and the meaning of not being in
Scotus); D. Carabine, The Unknown God: Negative Theology in the Platonic Tradition: Plato
to Eriugena. Louvain-Grand Rapids, Eerdmans Peeters-1995. W. Otten draws attention to
the text of Periphys. II, 99, 2369–2386 [597D–598A], where Scotus mentions sources such
as Augustine and Dionysius on the non-knowledge of God (‘The Shadow of the Divine:
Negative Theology and Anthropology in Augustine, Pseudo-Dionysius and Eriugena’, The
Heythrop Journal, 40 (1999), 438–455; p. 448). However, knowing that there is a tradition
from which Scotus draws, or that other authors made similar claims, does not help to
solve the epistemic paradox we have mentioned.
12 Periphys., I, 6, 86–90.
The Metaphysical System of Scotus Eriugena 59
conversely, angel properties (such as ‘invisible’ and ‘immortal’) do not fit man.
This cannot mean that a thing is not because it simply does not possess the
properties of something else, but it is unclear what else Scotus can mean. A
more understandable formulation is found at the end of the paragraph, where
Scotus states that a thing is insofar as it is known to itself or to higher things,
and it is not insofar as it is not known to lower things.
Material things known in space and time through observation of their gen-
erative processes are said to be in the third sense. Likewise, things not yet
formed are said not to be. Scotus invites us to consider, as an example, all fu-
ture men somehow implicit in the first man created by God and present ‘in the
nooks of nature’. 13
The fourth meaning, once again, has to do with knowledge. Things are said
to be insofar as they are the subjects of intellectual understanding, and con-
versely, those which are perceived as being subject to physical changes are said
not to be. Contrary to the first meaning, in this sense the being of things sub-
ject to the intellect is considered, while in the first, things that are not cannot
be the subject of intellectual understanding. As I will argue later, this means
that they are not abstract objects. In this case Scotus seems to contrast things
present to or perceived by the senses with abstract things such as their nature
or properties.
The fifth meaning has to do with how human nature, ‘which through sin
has destroyed the dignity of the divine image’,14 is to be considered. This is
the sense in which man is said to be before his fall and when he is ‘restored’ to
grace.
Scotus’ division lends itself to several different interpretations. One possible
reading which I find particularly interesting is the ‘logical’ one: certain things
subject to predication can be said to be or not to be according to whether the
predicate can be attributed to their names in true statements. For example,
from a certain point of view it is not true (but neither is it false) to say of
God the same things that we say of all creatures. Another reading is the epis-
temic one: three out of the five divisions considered above are about things
that are (and things that are not) insofar as they are ‘thought,’ in some sense,
by some faculty/function of the soul. Lastly, a theological (and/or mystical)
13 This issue is present in the De hominis opificio by Gregory of Nyssa (cf. PG 44, coll. 123–
256). On the importance of the text as one of Scotus’ sources, see C. Erismann in La
gènese, quot., 154–164.
14 Periphys., I, 8, 137–140.
60 chapter 3
Three are the motions of the soul: the first is from the soul [= intellect,
which Scotus regards as the essence of the soul], the second in accor-
dance with reason, the third with sensory faculty. The first motion is sim-
ple, it resides in the highest part of the soul, it lacks the ability to interpret
or understand what it tends towards […] The second motion is that by
which God, unknown in itself, is known as the cause of all things […]
this motion lies within the soul, which is moved in a natural way and by
means of it leads to its own scientific attention the natural reasons form-
ing all things; such reasons being realized ( factae) in the object known
only through cause, subsist eternally. […] The third one is a compound
motion, interacting with things that are outside the soul, it returns to the
soul the reasons of the visible things, deriving them from certain signs. 16
The soul, then, seems to know something, albeit in the limited sense of bring-
ing to its own attention the ‘reasons’ of things. These reasons are to be under-
stood both as natural and as supernatural causes involved in the processes of
creation and subsequent generation, and as specific and general properties.
But what does it means to know and what are the limits of (human) knowl-
edge? Some clarification on the sort of knowing which is attributed to the
second motion (i.e., to reason), is provided by the following text:
15 See M. Fournier, ‘Eriugena’s Five Modes’ (Periphyseon 443A–446A)’, The Heythrop Journal
2009, 581–589.
16 Periphys. II 63, 1469 -64, 1484: ‘Tres universales motus animae sunt, quorum primus est
secundum animum, secundus secundum rationem, tertius secundum sensum. Et pri-
mus quidem simplex est et supra ipsius animae naturam et interpretatione caret […]
Secundus vero motus est quo incognitum deum diffinit secundum quod causa omnium
sit […] et est motus iste intra animae naturam, ‘per quem ipsa naturaliter mota omnes
naturales rationes’ omnium formatrices quae in ipso cognito solummodo per causam ae-
ternaliter factae subsistunt operatione scientiae sibi ipsi imponit. […] Tertius motus est
‘compositus per quem quae extra sunt’ anima ‘tangens veluti ex quibusdam signis apud
se ipsam visibilium rationes reformat’.
The Metaphysical System of Scotus Eriugena 61
As we have said, the second motion of the soul defines God as a cause
within the limits of nature, i.e. it comes to know God – in Himself un-
known – as the cause of all things which are, [and also comes to know]
the primordial causes of all things as eternally grounded in Him and
through Him, and it impresses the intellectual apprehension of the intel-
ligible causes onto the spirit itself of which it is a motion, still not under-
standing what they are exactly (this exceeds the capacity of every motion
of the soul), but in such a way as to know universally why they are and
why their effects flow in an ineffable procession […] This motion, which
the Greeks call logos or dynamis, is called by us reason and virtue, and
arises from the first motion, which is the intellect.17
The network of relations between the motions of the soul is very complicated.
The intellect ‘fixes’ its objects in reason as species/genera and ‘observes’ them
as individual objects in the sensory faculty. Conversely, in the series of acts of
the soul, through the intellect, the reason acquires a kind of ability to unify the
multiplicity of experience, an ability ‘that comes from the simplest unity of its
principles.’ Scotus sums up his point of view in the following terms:
17
Periphys. II, 69, 1581–1592: ‘Motus animae secundus est, ut diximus, qui intra terminos na-
turae continetur ipsumque deum secundum causam diffinit, hoc est illud solummodo de
deo incognito quid est cognoscit quod omnium quae sunt causa sit, et quod primordiales
omnium causae ab eo et in eo aeternaliter conditae sunt, earumque causarum quantum
datur ei intellectarum ipsi animae cuius motus est cognitionem imprimit, non ut intelligit
quid substantialiter sunt (hoc enim superat omnem animae motum), sed ut universaliter
cognoscat quia sunt inque suos effectus ineffabili processione profluunt […] Et hic motus
qui a graecis logos vel dynamis, a nostris vero ratio seu virtus vocatur, et ex primo motu
nascitur qui est intellectus.’
62 chapter 3
way through sensory faculties as in the effects of the first causes. In other
words, everything that is understood by the senses as scattered in the ef-
fects, appears to reason as something that uniformly subsists in a unified
manner in the first causes.18
Reason does not know God, but comes to know Him as a cause. Similarly, it intel-
lectually apprehends causes, without knowing them, but in such a way as to
connect them to their effects. We could put things this way: objects of reason
and intellect are a kind of black boxes that produce effects independently of
or pursuant to other effects. We see and know – in a sense – the effects (or
their abstract details) and we have some idea that these effects arise for causes
to which, however, we do not have direct access. It seems that much of what
Scotus says has to do with the distinction between knowing things directly by
sense experience and knowing through clues and discursive reasoning.
It is quite clear that a believer knows many things about God, since he trusts
what is stated in the sacred texts and tradition of his confession. Similarly he
knows, in a sense, many things to which he has no direct access by reasoning or
by some testimony. Some additional explanation about intellectual knowledge
comes from the theory, or rather the metaphor, of abstraction:
‘The waters’ therefore ‘gather under the heaven in one place and the dry
land appears.’ If the contemplative soul did not first gather the floating
instability of matter and form that adheres to it as if they were surging
waves, then the substantial form, always established in the solidity of its
nature, would not appear to the eyes of the mind in a single place in the
intelligence (in fact, the intellectual understanding of matter and form
adhering to it is a unit).19
18
Periphys. II, 71, 1649–72, 1661: ‘Et ut breviter colligamus: quodcunque anima per primum
suum motum, qui est intellectus, de deo et primordialibus causis uniformiter et univer-
saliter cognoscit, secundo suo motui, qui est ratio, eodem modo uniformiter universali-
terque infigit. Quodcunque autem a superioribus per intellectum in ratione formatum
accipit, hoc totum in discretas essentias, in discreta genera, in diversas formas, in mul-
tiplices numeros per sensum inferioribus effectibus distribuit, et ut apertius dicam,
quodcunque anima humana per intellectum suum in ratione sua de deo deque rerum
principiis uniformiter congnoscit, hoc totum per sensum in causarum effectibus multi-
pliciter intelligit. Et iterum, totum quod per sensum multipliciter sparsum in effectibus
intelligit, per rationem in causis unum uniformiter subsistere perspicit.’
19
Periphys. III, 120, 3462–3469: ‘‘Congregentur’ ergo ‘aquae quae sub caelo sunt in locum
unum, et appareat arida.’ Nisi enim contemplativus animus fluctuantem materiae ad-
herentisque ei formae instabilitatem, veluti quosdam aestuantes fluctus, in unum
The Metaphysical System of Scotus Eriugena 63
Just as the waters cover the earth entirely, so that it does not appear to
the senses, so the mutability of corruptible bodies and the innumerable
multiplication of what subsists prevent the eyes that contemplate the na-
ture of things from detecting the stability of their form, so that it cannot
be considered independently of the bodies in which it is. Just as when,
once the waters have flowed back and the waves have returned to their
beds, the bare, dry and solid shores clearly appear, so it is with the mind,
impeded in its attitude of contemplation of the incorruptible natures by
the instability of corruptible things: the beautiful and unchanging stabil-
ity of essential forms and species in their genera are pleasing to the soul’s
gaze. The stability of essential forms is said, not without reason, to be
arid, as it is stripped of and freed from all covering of accidents.20
This ‘drying out’ of the images deposited in the memory expresses the idea that
the object considered by the rational soul is obtained not directly but by the
subtraction of features. There is, of course, a difference between what the soul
abstracts from images and what corresponds to abstract notions. Scotus has a
peculiar theory according to which the object in the soul is more important
than what corresponds to it out of the soul.21 One possible reason for this is
that Scotus considers the soul something nobler than other created things, and
would accept a maxim such as ‘that which resides in a nobler element must
be nobler too.’ Whatever the case, the concepts of incorruptible natures, sub-
stantial forms, and species appear to be obtained by abstraction from sensible
things. Can one assert the same about the remaining incorruptible natures? In
one sense, one cannot, because no one has seen an angel; in another sense one
can, because in principle seeing angels (such as in the case of angelic appari-
tions) and God (in the person of Christ) is conceivable.
The best way to explain the link between conceivable and inconceivable
things seems to be to consider what it is not as an object that does not match
what is generated in the soul through abstraction. Being connected by a causal
link – in some sense of the word – to physical objects does not entail knowabil-
ity in the strict sense. In other words ‘knowing’ is said in three ways: a) knowing
directly, by acquaintance, b) knowing indirectly via an abstractive process, and
c) knowing discursively, by attributing to an object properties by faith or by
reasoning. Typically for this last sort of knowledge there are causes which are
unknowable in themselves, but whose effects are directly knowable.
But what about essences/substances? If the essence is a kind of Aristotelian
substance that remains unchanged through accidental changes, there should
be a sense in which such an essence is not an object of knowledge through
abstraction. If we understood the objects of abstract knowledge as properties
common to several individuals, or common aspects, or natures, substances
would be unknowable, since they are not common properties, aspects or na-
tures, but that which has properties, presents an aspect, or reveals a certain
nature. If, on the other hand, one considers abstraction to be a process that not
only separates properties from their bearers, but also items from properties,
then species and genera can be seen (in a categorial chain) as objects separate
from the properties possessed by the objects ‘under’ them. Scotus seems to
claim that essences are not knowable because they are not obtained through a
process of subtraction of properties.
For example, I look at Socrates and I see his properties and say that he is a
philosopher. Starting again from the observation of Socrates, I consider what
does not have all the individual properties owned by Socrates, but only has
a subset of such properties, and call it man. While identifying this object, I
observe that it has the property of being an animal, etc. If Scotus considered
the process of abstraction in a similar way, he may have thought that essences
could not be known as objects deprived of certain properties, but rather as
causes for which objects have the properties that they have. In conclusion,
Scotus identifies a deductive, empirical sort of knowledge that enables us to
define abstract objects, as well as a hypothetical reasoning which assumes the
The Metaphysical System of Scotus Eriugena 65
existence of causes for certain natural processes. What is not known in this
second way can be equally hypothetically considered as endowed with proper-
ties. This explains why we can talk about essences and God, or believe what
others have said about them.
22 Cf. Periphys. I, 32, 887–895. See E. Hochschild Paige, ‘Ousia in the Categoriae decem and
the Periphyseon of John Scottus Eriugena,’ in M. Treschow – W. Otten – W. Hannam (edd.),
Divine Creation in Ancient, Medieval, and Early Modern Thought. Essays Presented to the
Rev’d Doctor Robert D. Crouse, Brill, Leiden 2007. pp. 213–222. On Categoriae decem as a
source of Scotus: J. Marenbon, ‘John Scottus and the Categoriae decem,’ cit. above.
23 G. D’Onofrio insists in a number of works on the importance of Boethius’ theological
writings as a source for Scotus, one of which is ‘On ‘Magnificus Boetius’: An Investigation
into the Presence of the Sacred and the Opuscola Sacra and Consolatio in Eriugena’s work,’
in W. Beierwaltes (ed.), Eriugena. Studien zu seinen Quellen, Carl Winter, Heidelberg 1980;
pp. 189–200; ‘Agli inizi della diffusione della Consolatio e degli Opuscula sacra nella scuo-
la tardo-carolingia: Giovanni Scoto e Remigio di Auxerre,’ in L. Obertello (ed.), Atti del
Congresso Internazionale di Studi boeziani (Pavia, 5–8 ottobre 1980), Herder, Roma 1981.
pp. 343–354; ‘Dialectic and theology: Boethius’ Opuscula sacra and their early medieval
readers,’ Studi Medievali 27 (1986), 45–67. As for De Inst. Arith., it is interesting to note
that Scotus cites the crucial passage of the De Inst. Arith. where Boethius presents a non-
Aristotelian list of ‘things that are’ (ed. Friedlein, Teubner, Leipzig, 1867, p. 7, 26–8, 13).
24 Periphys. I, 97, 3016–3017: ‘Significativa rerum vocabula, sive substantiarum sint, sive ac-
cidentium, sive essentiarum.’
25 Periphys. I, 111, 3485–3486: ‘(Deus) plus est quam esse, et causa omnis esse et essentiae et
substantiae.’
66 chapter 3
Not only the primordial causes are founded in the Word, but also their
effects, places, times, essences and substances, i.e. the highest genera, the
common forms, the lowest species instantiated in the individuals.26
The substances of created things, the essences and the reasons […]
proceed and descend from primordial causes.27
Each essence and substance is not and does not subsist, but for super-
substantial and superessential goodness.28
26
Periphys. III 74, 2109–2117: ‘Non solum primordiales causas, verum etiam earum effectus
et loca et tempora et essentias et substantias, hoc est genera generalissima et generales
formas et species specialissimas per individuos numeros.’
27
Periphys. V, 38, 1148–1151: ‘Rerum conditarum substantiae et essentiae et rationes ex pri-
mordialibus causis […] procedunt atque descendunt.’
28
Periphys. V, 62, 1996–1998: ‘Omnis essentia et substantia non aliunde et est et subsistit nisi
ab ea superessentiali et supersubstantiali bonitate.’
29
Periphys. V, 62, 2001–2003: ‘Nulla enim essentia vel substantia est a divina bonitate condita
quae non aeternaliter et incommutabiliter permaneat.’
30
Periphys. V, 77, 2437–2449: ‘omnia siquidem quae […] sensibus soccumbunt, non ipsae
res substantiales vereque existentes, sed ipsarum rerum vere existentium quaedam tran-
sitoriae imagines et resultationes intelligenda sunt […] sicut imagines vocum umbraeque
corporum per se non subsistunt, quia substantia non sunt, sic corpora ista sensibilia ve-
luti rerum subsistentium quaedam similitudines sunt.’
The Metaphysical System of Scotus Eriugena 67
It must then be stated clearly, by all experts in physics, that the nature
and substance of bodies is incorporeal. 31
31 Periphys. V, 185, 6024–6026: ‘Omnibus physicis sapientibus dum sit diffinitum naturam
substantiamque corporum incorpoream esse.’
32 Scotus often prefers to use the Greek ‘ousia’ instead of the Latin ‘essentia’; it is neverthe-
less clear that the translation from the Greek should be ‘essentia’ and not ‘substantia’,
despite the fact that the concepts of essence and substance sometimes seem to overlap.
In the English translation I will usually use ‘essence’, but the Greek will be maintained in
quotations. ‘Substantia’, of course, is translated as ‘substance’.
33 Periphys. I, 5, 68–73: ‘Quicquid autem in omni creatura vel sensu corporeo percipitur, seu
intellectu consideratur, nihil aliud est nisi quoddam accidens incomprensibil<e>, ut dic-
tum est, uniscuiusque essentiae. Nam aut per qualitatem, aut quantitatem, aut formam,
aut materiem, aut differentiam quandam uniuscuiusque essentiae.’ The critical edition
has ‘incomprensibili’, and PL ‘incomprensibile’; ‘incomprensibilis’ is in the versio IV of the
critical edition. However, in the same versio there is ‘unicuique essentiae’, whereas in PL
there is ‘unicuiusque’. My reading puts together the variants in versio IV and PL.
34 Periphys. I, 71, 2171–2175. C. Kavanagh believes that the contribution of Maximus the
Confessor is important. He also quotes a helpful text on unknowability (Periphys. I,
471B–472B) in which Scotus speaks of periochae or circumstances (circumstantes)
68 chapter 3
Scotus’ position is somewhat paradoxical: on the one hand he states that es-
sences, among other immaterial things, are incomprehensible, yet he provides
us with a lot of information about them. The unknowable is equated with the
non-definable. With regard to the essence-substance, understood as the high-
est genus (from a categorial perspective) it is clearly not possible to find a defi-
nition, but the specific essences-substances (such as /man/) should be defined
in the traditional way.
The following text addresses the (lack of) definability of essence:
Each ousia is simple, and does not admit any composition of matter and
form, as it is an inseparable unity. It is therefore unreasonable to concede
that any ousia is a mortal body. […] Every ousia, though it is considered
to be formed by essence [probably the supreme genus /essence/] and es-
sential difference (in fact no essence can fail to have such a composition),
[…] this composition itself, which is only known through reason, does
not appear to be caused by any act or operation, and is therefore to be
judged as simple bareness. In order to better grasp that the ousia, i.e. the
essence, is incorruptible, see the De divini nominibus by Dionysius the
Areopagite.35
which ‘reveal’ the presence of an essence. See C. Kavanagh, ‘The Influence of Maximus
the Confessor on Eriugena’s Treatment of Aristotle’s Categories’, American Catholic
Philosophical Quarterly, 79 (2005), 567–596, 581–83 (this particular issue is dedicated to
Scotus).
35
Periphys. I, 66, 2027–2040: ‘Item omnis ousia simplex est, nullamque ex materia formaque
compositionem recipit, quoniam unum inseparabile est. Nulla igitur ousia corpus mor-
tale rationaliter conceditur esse. […] omnis ousia quanquam intelligatur ex essentia et
essentiali differentia composita esse – hac enim compositione nulla incorporea essen-
tia potest carere – […] ipsa tamen compositio, quae sola ratione cognoscitur, nulloque
actu et operatione fieri comprobatur, rationabiliter simplicitas esse iudicanda est. Ut
autem firmius cognoscas uosian id est essentiam incorruptibilem esse, lege librum sancti
Dionysii Aeropagitae de divinis nominibus.’
The Metaphysical System of Scotus Eriugena 69
I’m not entirely certain how the ousia can be defined into genera, species
and atoms, having previously concluded that it is not understood by the
intellect and does not fall under the senses.37
The Master’s reply is supposed to provide the ultimate solution to this problem:
No one can define or say what ousia is in itself; however it can be de-
fined by the things adhering to it, and without which it cannot be, I mean,
without place and time. In fact, every ousia created from nothing is posi-
tioned in space and time […] So you cannot define what essence is, but
the reason why it is, and only the reason can be given (quia est), not the
what (quid est), according to the time and fulfillment of other accidents
that are considered – as we have said – internal or external to the essence
[…] whoever said this in general of any ousia, be it highest, lowest or in-
termediate, did not express himself improperly.38
36 Here I would like to mention a comment by W. Uzgalis in his article on Locke for the
Stanford Encyclopedia: ‘These kinds of reflections led him [Locke] to the relative and ob-
scure idea of substance in general. This is that ‘I know not what’ which is the support
of qualities which cannot subsist by themselves. We experience properties appearing
in regular clumps, but we must infer that there is something that supports or perhaps
‘holds together’ those qualities. For we have no experience of that supporting substance.’
(W. Uzgalis, ‘John Locke’, in E. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter
2010 Edition): http://plato.stanford.edu/archives/win2010/entries/locke/). An examina-
tion of the thesis according to which substance is an aggregate of properties can be found
in MJ Loux, Substance and Attribute, 112–119.
37 Periphys. I, 63, 1906–1910: ‘Non mihi plane patet quomodo ousia sive in generibus sive in
speciebus sive in atomis definiri possit, cum in praedictis rationibus huius libelli conclu-
sum sit eam nulli sensui corporeo, nulli intellectui comprehensibilem esse.’
38 Periphys. I, 63, 1911–1922: ‘Ousian per seipsam diffinire et dicere quid sit nemo potest. Ex
his autem quae inseparabiliter ei adherent, et sine quibus esse non potest, ex loco dico
et tempore – omnis enim ousia de nihilo creata localis temporalisque est – […] Ousia
70 chapter 3
The concept of ‘reason why’ plays a role in some of the explanations that
Scotus provides on the unknowability and non-definability of essences. The
distinction between ‘the reason why it is’ and ‘what it is’ may depend on the
Boethian one between id quod and esse. One can detect a connection between
the pairs id quod-esse (Boethius), quod est-quia est (Scotus), and id quod-quo
est (Gilbert of Poitiers); they all seem to insist on the distinction between the
bearer subject and inherent properties that exert a causal power of some kind.
After an analysis of the concept of essence in Book 1 (the central pages on
the issue), the disciple draws the Master’s attention to the distinction between
form and matter, and the question whether essential forms are of the same
type as those which shape matter. The Master first proposes a general presen-
tation of this theme:
As usual, this text raises problems. What does it mean that the forms are in es-
sence? Does it mean that from a certain point of view they are essences, or that
they are a kind of essential property or properties of the essence? Considering
itaque nullo modo diffinitur quid est, sed diffinitur quia est. Ex loco namque, ut diximus,
et tempore, accidentibusque allis, quae sive in ipsa, seu extra, intelliguntur esse […] hoc
generaliter de omni ousia sive generalissima, sive specialissima, sive media non incon-
grue quis dixerit.’
39 I have translated in ousia, in qualitate literally. Note that later (cf. infra col. 496) Scotus
apparently speaks in the same sense of the forms of essence (ousiae) and of quality
(qualitatis).
40 Periphys. I, 72, 2217–2225: ‘Formarum aliae in ousia, aliae in qualitate intelliguntur; sed
quae in ousia sunt, substantiales species generis sunt. Nam de ipsis genus praedicatur,
quia in ipsis subsistit. Genus nanque, ut saepe diximus, totum in singulis suis formis est,
quaemadmodum et singulae formae unum in suo genere sunt. Et haec omnia, id est gen-
era et formae, ex uno fonte ousiae manant […] Formae vero quae qualitati attribuuntur,
in naturalibus corporibus proprie formae, in geometricis autem figurae vocantur.’
The Metaphysical System of Scotus Eriugena 71
them to be substantial species would support the first hypothesis, but would
entail an unconventional concept of form, although Boethius does sometimes
seem to take such a stance.
Let me start with some examples of forms ‘in essence’ and forms ‘in quality’:
The order and position Form as figure, the figure We wonder what the
of the parts, such as the being the quantity of space, essence of this body is,
upright form of man.a the organization of the lines i.e. its substantial form,
That form dependent on and the quality.e whether it is human
quality and quantity befits or equine. ‘Human’ or
natural bodies.b ‘equine’ do not refer to
Shapes and colors of animal bodies but to
physical objects.c their substantial forms.f
The body is composed of Forms created (or not)
the four elements and the and creating (or
form which compose and non-creating); (‘form’ is
shapes them.d used here as a synonym
for ‘nature’).g
Division of the
essences and genera
into forms and
numbers. Reunification
(Recollectio) of
individuals into forms
and genera.h
It is worth noting that the examples of natural forms in the quality do not seem
to be of the same sort. On the one hand Scotus considers order and position of
the parts, such as the upright form of man, on the other the forms generating
physical bodies. As for substantial forms or forms in essence, things are even
more complicated. The possibilities at stake, at least according to what the ex-
amples suggest, seem to be: a) forms as species of the genus /essence/, b) forms
as elements of some lowest species, c) forms as related to the abstract denota-
tions of the names of species (humanitas vs homo), and d) forms as properties
of essences, specific differences or sum of differences.
The two interlocutors, protagonists of the De Divisione, first appear to agree
on the distinction between qualitative and essential forms:
N: Nothing is clearer: one form is in the ousia not as accident but as some-
thing next to it, and the other (the qualitative form = ex qualitate) pro-
duces a perfect body when it is added to quantity, isn’t it so? A: I can
already see what you are getting at. 41
N: As you understand what I’m getting at, then tell me if this division of
forms into two different species of the genus (/form/), i.e. forms of ousia
and forms of quality, seems right to you. A: I think so, although I have not
accepted it without reservations. You would be more likely to convince
41
Periphys. I, 74, 2265–2269: ‘Nonne luce clarius est, aliam esse formam quae in ipsa ousia
non ut accidens, sed ut idipsum ei perspicitur, aliam quae ex qualitate quantitati adiuncta
corpus efficit perfectum? A: Iam video quo tendis.’
The Metaphysical System of Scotus Eriugena 73
The doubt expressed by the disciple is compatible with the Aristotelian per-
spective, according to which the form, which is the essence of a living body,
determines certain matter to be what it is. Form intended in such a way is the
substance of the material body itself. Indeed, the disciple declares that he be-
lieves essence to be the most likely cause of the constitution of the body, rather
than quality, which joins matter together with quantity. The teacher expresses
his admiration for the disciple’s cleverness; after pointing out that the qual-
ity and quantity in question are incorporeal entities, he offers the following
summary:
Is it not, therefore, more likely that anything made with quantity and
quality, i.e. that is made qualitatively and quantitatively, has no other
cause of its own constitution than the ousia in which – as it is proved –
quality and quantity are inherent as first and main accidents […]? I do
not see why anything that comes from those things that are in the source
cannot be referred to the same origin. […] quality and quantity together
give rise to a qualitative and quantitative something […] (quality and
quantity) are originally observed in our bodies, and we contemplate
them outside of these bodies because of a higher consideration of the
ousia, which is the origin of substantial forms.43
I believe Scotus is providing reasons to believe that the qualitative forms are
somehow connected to the essence; this is what his response to the disciple’s
42
Periphys. I, 75, 2308–2315: ‘N: Quoniam igitur video te intentionis nostrae finem acute
prospexisse, dic oro num tibi haec divisio formarum in duas species diversi generis ousiae
videlicet atque qualitatis rata esse videtur necne? A: Rata quidem, verique similis, licet
non sine animi mei scrupolo hanc rationabilem esse concesserim. Etenim facilius mihi
suaderes formam substantialem materiae adiectam naturale corpus constituere, quam
qualitativam.’
43
Periphys. I, 76, 2339–2362: ‘Num itaque probabiliter dicendum est omne quod quanti-
tate et qualitate conficitur, hoc est quantum et quale, non aliunde constitutionis suae
causam accipere nisi ab ipsa ousia, cui naturaliter ipsa quantitas et qualitas veluti prima
accidentia maximaque probantur accidere […]? Quicquid enim ex his quae in fonte sunt
progreditur, cur non ad ipsum fontem referatur non habeo […] Quantitas siquidem et
qualitas quantum et quale inter se coniungunt […] Hec, ut diximus, in corporibus nostris
primordialiter inspiciuntur. Extra vero haec altiori consideratione ousian quae est for-
marum substantialium origo contemplamur.’
74 chapter 3
‘Aristotelian’ objection would consist in. The qualitative and quantitative ‘con-
crete’ forms are characterized as those particular accidents (whiteness, height,
shape, etc.) which give rise to individual bodies by bonding with the essence
and substantial forms. Scotus does not want to follow the disciple in arguing
that substantial forms give rise to bodies; on the other hand, however, he does
not want to stray from Aristotle’s doctrine by saying that accidents and not es-
sences are the cause of physical bodies. His conciliatory solution is to consider
qualitative forms as generative inasmuch as they are grafted onto substantial
forms and ultimately onto essence itself.
One might wonder whether qualitative abstract forms are subject to change
just as the concrete ones are. In Book 3 Scotus speaks of simple qualities and
quantities that are invisible and inaccessible to the senses,44 evidently re-
ferring to the abstract forms which adhere to the essence directly. Later, in
Book 5, he considers the material form of the subsisting body to be immu-
table and incorruptible, whereas the shapes added to the bodies are subject
to changes:
Gregory of Nyssa claims that not only is the form of the soul made in
the image of God, but also the natural form of the body, which imitates
the image of the soul and remains immutable and incorruptible. What is
added to the natural body by concretion of the elements, and to the soul
by the baseness of irrational motions, is always immersed in change and
corruption.45
In this case the lectio facilior is to understand the form of the body as a specific
difference. The lectio difficilior is that even forms in matter which give physical
bodies their appearance present themselves as immutable and abstract when
examined with the mind’s eye. One possible reason for preferring the second
reading has to do with the problem of individuation. In order to have indi-
vidual essences it is necessary to identify special qualities of the essence that
44
Periphys. III, 65, 1861–1863: ‘(omnia corpora composita et solubilia) ex qualitatibus enim
et quantitatibus simplicium et invisibilium corporum sensibusque incomprensibilium
efficiuntur.’
45
Periphys. V, 19, 537–20, 543: ‘Nyseus astruti quod non solum forma animae ad imaginem
dei facta, vero etiam naturalis forma corporis, quae imaginem animae imitatur, semper
incorruptibilis et incommutabilis permaneat. Quicquid vero naturali corpori ex concre-
tionibus elementorum, et animae ex sordibus irrationabilium motuum, superadditum
est, in fluxu et corruptione semper est.’ On the contrast between the material and spiri-
tual body, see also ibid. p. 177 et seq.
The Metaphysical System of Scotus Eriugena 75
He called high things such as /man/, /horse/, /lion/ eide, almost as if they
were part of the genus and forms of things; then there are higher genera
[…] from which parts or forms arise.47
Species on the other hand – which some also call form – is what […] is
predicated in the essence according to number […] like /man/.48
The author of the Categoriae decem seems to contrast the species, the forms
of things, and the genera, by providing the lowest ones (/man/, /horse/, /lion/)
as an example of species-forms. Actually, there is no reason to differentiate
entities like /man/ from genera such as /animal/ or /substance/. If there is
some distinction, it most likely has to do with the difference between substan-
tial forms that specifically determine a singular thing to be what it is (such
as ‘being a man’ for Socrates), and other forms or generic entities signified by
common nouns of substance. The specific differences that would be the first
candidates for the role of substantial forms are in some way set aside. I believe
that this difficulty can easily be resolved if we think of things such as humanity
as the sum of differences, or a ‘condensed’ difference that allows us to identify
a certain species in a genus: for instance, the genus ‘animal’ is distinguishable
in animals having humanity, equinity, etc..49
In order to complete the explanation, it may be useful to (re)consider some
of Boethius’ texts. The first text is on the meanings of ‘species’. Boethius uses
the term ‘form’ to talk about a second meaning of ‘species’ not considered by
Porphyry: ‘(in addition to the form-appearance of any thing) there is that other
kind of substantial form which is called humanity, which is not subordinate
to /animal/, but shows the substance as well as quality.’50 Boethius introduces
here a kind of entity that is difficult to place in the Aristotelian classification:
namely, the meanings of abstract nouns derived from the names of species/
genera, such as ‘animality’ from ‘animal’. A second text (already cited in part) is
taken from the theological works:
As for the fact that other forms (other than the divine form or those with
which theology is concerned) are subject to accidents, such as humanity,
[it should be noted that] pure form does not receive accidents for the
reason that it is, but because matter is subject to it, and in fact while the
matter subject to humanity receives some accident, humanity itself also
seems to receive it. Actually, the form which is without matter cannot be
a subject, nor be concerned with matter; if it were so, it would not be a
form but an image. These forms that are in matter and that produce the
body come from those which are without matter. We commit abuse by
calling the remaining ones that are in bodies forms, when we should call
them images.51
49 One might wonder whether the substantial form ‘animality’ exists. If what we have said is
true, there should be at least as many substantial forms as species and genera. If the lack
of examples is not due to a mere accident, as would certainly seem to be the case, one
reason may be that substantial forms are a kind of ‘super-difference’ dividing genera.
50 In Isag. II, 200, 7–16, cit.
51 De Trinitate, ed. cit, 10, 44 -12, 54: ‘Nam quod ceterae formae subiectae accidentibus sunt
ut humanitas, non ita accidentia suscipit eo quod ipsa est, sed eo quod materia ei subiecta
est; dum enim materia subiecta humanitati suscipit quodlibet accidens, ipsa hoc suscip-
ere videtur humanitas. Forma vero quae est sine materia non poterit esse subiectum nec
vero inesse materiae, neque enim esset forma sed imago. Ex his enim formis quae praeter
materiam sunt, istae formae venerunt quae sunt in materia et corpus efficiunt. Nam ce-
teras quae in corporibus sunt abutimur formas vocantes, dum imagines sint’.
The Metaphysical System of Scotus Eriugena 77
The medieval discussion on the issue of universals focuses on the status of spe-
cies and genera. It is therefore of some interest to document Scotus’ opinions
on this subject. The concept of species/genus is a relative concept, and it sup-
poses a given set of objects and an order relation defined on it. The standard
example for the logicians is the so-called Porphyrian tree, where the object-
nodes are substances (a set of selected substances) and the branches represent
the order relation of the being-said-of. Scotus actually speaks several times of
species and genera, evidently considering a kind of relationship, but its under-
lying set varies depending on the context.
When discussing the creation of the world and the degree of perfection
of creatures, Scotus considers a division which begins with God and ends in
the species /man/, which is then in turn divided into /male/ and /female/.
Elsewhere he describes a process for the production of genera, forms and spe-
cies ‘from unknown hidden crevices of nature.’ The following passage from
Book 4 offers an example of how theological, creational, logical, and poetic
issues are blended together:
There exists some very general nature, common to all things, created by
the unique principle of all things, from which, as from a bountiful source,
corporeal creatures derive from hidden crannies, almost like a stream,
and they gush in the various forms of individual things […] Man, who
was formed on the sixth day, is placed in the number of animals, which
are included in a single genus […] ‘produces the earth the living soul.’
Since [the biblical text] adds ‘of its kind,’ it clarifies how all animals are
contained in this order.52
52
Periphys. IV, 14, 328–345: ‘Est enim generalissima quaedam atque communis omnium na-
tura, ab uno omnium principio creata, ex qua veluti amplissimo fonte per poros occultos
78 chapter 3
The art which the Greeks call dialectics, defined as the science of good ar-
guing, first of all concerns the ousia, i.e. its proper principle, from which
each division and multiplication of those things which it deals with de-
rives; [dialectics] starts from the higher genera and from intermediate
ones, descending to the very particular forms and species, and once again
by the rules of multiplication, by the same degrees through which it de-
scended, returns to the very ousia it started from.53
Of course, here too we find a shift to a plane that is foreign to dialectical discus-
sions: after the descent, the art of dialectics – in its personification – returns
to the source from which it moved in order to rest and enjoy the contempla-
tion of the essence. Similar passages are not very useful in reconstructing the
history of universals, apart from documenting a broad realist point of view on
the problem. Two aspects seem to me, however, to be worthy of attention: the
relation of being-said-of, and what Scotus says on the concept of universal.
Let us start from the discussion contained in the first book on the canonical
distinction between to be-said-of and to be-in.
This topic, discussed at the end of a brief presentation of the categories,
does not permit much speculation, since the whole tradition of the commen-
tators (including Categoriae decem) simply reiterates the Aristotelian distinc-
tion; Scotus, however, offers his own opinion:
being a subject and saying of a subject do not differ in any way. If indeed,
as some say, Cicero is a subject and a first substance, and ‘man’ is [said] of
a subject, being a second substance, what is the difference according to
nature except that the former is numerically one, and the latter is one in
species, since species is nothing more than a numerical unity, and num-
ber nothing more than a plurality of species? If, therefore, species as a
whole is one and individual in numbers, and numbers are one and indi-
vidual in species, I do not see what difference in nature there is between a
subject and [what is said of] a subject. Similarly for accidents of first sub-
stances. In fact, what is said in the subject is nothing other than what is in
and is said of the subject. Discipline – for example – is one and identical
in itself and in its species and numbers; any particular discipline, only
in a subject according to dialecticians, is no different from the general
discipline which for them is said of and is in a subject.54
54
Periphys. I, 42, 1224–1242: ‘Et hoc dignum quaesitu video. Nam iuxta dialecticorum opin-
ionem omne quod est aut subiectum, aut de subiecto, aut in subiecto et de subiecto est.
Vera tamen ratio consulta respondet subiectum et de subiecto unum esse, et in nullo
distare. Nam si, ut illi aiunt, Cicero subiectum est et prima substantia, homo vero de
subiecto secundaque substantia, quae differentia est iuxta naturam, nisi quia unum in
numero, alterum in specie, cum nil aliud sit species, nisi numerorum unitas, et nil aliud
numerus, nisi speciei pluralitas? Si ergo species tota et una est individuaque in numeris,
et numeri unum individuum sunt in specie, quae, quantum ad naturam, distantia est
inter subiectum et de subiecto, non video. Similiter de accidentibus primae substantiae
intelligendum. Non aliud est enim, quod in subiecto dicitur, et aliud, quod in subiecto
simul et de subiecto. Nam disciplina, ut exemplo utar, una eademque est in seipsa, et
in suis speciebus numerisque. Non aliud igitur uniuscuiusque propria disciplina, quae a
dialecticis in subiecto dicitur solummodo, et aliud generalis disciplina, quae ab eisdem in
subiecto et de subiecto vocatur.’
80 chapter 3
D.: Must we consider two substances of man, a general one in the primor-
dial causes, and a special one in their effects? Master: I have not said that
55 In the preamble to the second commentary on Porphyry (see above), Boethius proposes
the following argument: all that exists is numerically one, genera and species are com-
mon, and what is common cannot be one; thus genera and species do not exist. On the
other hand, Boethius admits – as does Scotus in the text we are commenting on – that the
genus is one in its individual species. For the next period, see what we say on the concept
of thing in Abelard.
56 The relationship to the Isagoge is, however taken into account by scholars who claim to
have found more evidence than I have been able to trace; see for instance: C. Erismann,
‘Processio id est multiplicatio. L’influence latine de l’ontologie de Porphyre: le cas de Jean
Scot Érigène’, Revue des Sciences Philosophiques et Théologiques 88 (2004), 401–460; ‘The
Logic of Being: Eriugena’s dialectical ontology’, Vivarium 45 (2007), 203–218 (in particu-
lar p. 204). The same author notes the total absence of quotations from Porphyry in the
Periphyseon, and that the only mention of Isagoge in the Annotationes in Marcianum
may come from a textbook summary. However, there are – in Erismann’s opinion –
independent reasons for thinking that Scotus had Porphyry’s text in mind. In fact accord-
ing to Erismann ‘il s’agit de montrer comment l’Erigène aide par sa lecture de Maxime le
Confesseur et du Pseudo-Denys, a su ‘exploiter’ le néoplatonisme ‘résiduel’ du texte de
Porphyre’ (‘Processio..’, p. 403). Among these reasons, what seems to me to be the main one
is the presence of the Isagoge, together with Categoriae decem, in scholastic manuscripts
from the ninth and tenth centuries. See, in this regard, Marenbon, ‘John Scottus and the
Categoriae decem’ cit., 118.
The Metaphysical System of Scotus Eriugena 81
there are two, but rather one differently conceived. In a way the human
substance is considered according to its condition in the intellectual
causes, in another way according to generation in the effects. There it
is free from any mutability, here is full of mutability, there it is simple
and free from any accident, and escapes any consideration by the created
intellect, accepting the mind’s gaze here, acquires a certain composition
from quantity, quality and other forms which are understood as being
inherent in it.57
What does Scotus mean here? If he is talking about second substances, he may
be following Boethius’ denial of the difference between general and particular.
The reader will remember that in the enigmatic finale of the commentary on
Porphyry, among other things Boethius sets out his own version of the Neo-
Platonic theory of the three modes of being of the universal, claiming that the
universal is no different from the particular. Another possible interpretation
comes from the later identity theory, formulated between the eleventh and
twelfth centuries to avoid some of the difficulties associated with extreme real-
ism. Both of these possible readings seem to me to go beyond the text; I believe
that Scotus’ position is simply one of extreme or naive realism that considers
the universal to be something numerically one which, in some sense, is in sev-
eral things diversified by accidental features. The discipline does not consti-
tute a counterexample, since it is identical in several bodies and diversified by
degrees and measures. I believe that Boethius’ theological works (Rand 10–12)
are still the key source for this claim.58
57 Periphys. IV, 44, 1175–1185: ‘A: ‘Duas igitur substantias hominis intelligere debemus, unam
quidem in primordialibus causis generalem, alteram in earum effectibus specialem.
N: Duas non dixerim, sed unam dupliciter intellectam. Aliter enim humana substantia
per conditionem in intellectualibus perspicitur causis, aliter per generationem in effec-
tibus. Ibi quidem omni mutabilitate libera, hic mutabilitati obnoxia; ibi simplex omni-
busque accidentibus absoluta omnem effugit conditum intellectum, hic compositionem
quandam ex quantitatibus et qualitatibus, caeterisque, quae circa eam intelliguntur, pa-
tiens mentis recipit intuitum.’
58 In this connection, Marenbon notes: ‘the species [according to Scotus] is ‘whole and one
in its numerous members, and these members are one individual in the species.’ The only
way I can see of making sense of this position is to suppose that Eriugena meant by the in-
dividual of such and such a class simply the such-ness, by which he belongs to that class.
He would distinguish implicitly between the particular man, as he might be perceived
[…] and the individual member of the species, Man, who is simply humanity, on account
of which he is a man.’ (‘John Scottus’, cit., 128) . If Marenbon is talking about individual
essences as distinct from the essences /man/ and /essence/, and from individual men, I
82 chapter 3
The texts we have just examined allow for the slight possibility of mitigat-
ing extreme realism. In certain other passages, the realist thesis of the unity of
the universal in its particulars is more clearly articulated and is not subject to
misunderstanding. Let us consider a passage from Book 1, where Scotus states
that the specific essence is unique in singular men:
The ousia of all men is one and the same, in fact, everyone participates
in the essence, and for this reason it is common and not specific to any-
one […] The ousia is divided into genera and species, while the body –
understood as a certain whole – is divided into parts. The body in its parts
is not yet a whole […] while the ousia is entire in its individual species
and forms […] The ousia is not greater in all men than in a single man,
nor less in one than in all.59
agree: in the texts we have read Scotus appears to retain such a distinction. I do not know
why Marenbon feels the need to call these individual essences ‘man-ness’, making them
a sort of abstract property, something not substantial but qualitative. On this point, see
what we say above about the first substances in the paragraph on the essence.
59
Periphys. I, 68, 2089–70, 2134: ‘Omnium hominum una eademque ousia est. Omnes enim
unam participant essentiam, ac per hoc, quia omnibus communis est, nullius propria est.
[…] Nam ouisia in genera et species dividitur; corpus vero veluti totum quoddam in partes
separatur. Item corpus in partibus suis totum non est. […] E contrario autem ouisia tota
in singulis suis formis speciebusque est; […] ousia non est maior in omnibus hominibus,
quam in uno homine, nec minor in uno homine, quam in omnibus hominibus.’
60
Periphys. V, 28, 824–832: ‘Nam et substantia corporis profecto intelligibilis est, atque
ideo non est incredibile, nec rationi obsistens, si intelligibiles substantiae sibi invicem
The Metaphysical System of Scotus Eriugena 83
Other examples concern numbers that exist as unities in the monad and the
points that make up the line. It is rather curious that Scotus’ mathematical
examples intersect the line established by Boethius and Nichomacus to dis-
tinguish unconnected essences from continuous ones, and it is all the more
curious because the Institutio is definitely known to Scotus: lines are in fact
geometric entities listed by Boethius as continuous essences, numbers as dis-
crete essences.62 The example of light is a little more complicated, as it is not
clear how many and which parts merge into the diffused light.63
The standard definition of ‘universal’ excludes the possibility that the uni-
versal is a whole whose parts are the species or individuals. However, in the
De Divisione Boethius says that the universal can be considered as a whole sui
generis, and in the Categoriae decem the species are sometimes called parts of
the genus.64 The examples proposed do not all lead in the same direction: the
monad is, in some sense, in the numbers, individual lights in diffused light,
points in the line, metals in the alloy. Changing the example, also changes the
way of understanding the relationship between ‘component’ and ‘composed’.65
How much credit Scotus gives his metaphors may not be entirely clear, but
there is no doubt that the the realist thesis of the unity of the universal is ac-
cepted. This is confirmed by a text that seemingly acts as a seal affixed to the
discussion of universals:
64 Cf., for instance, the text of Categoriae decem quoted above: 133–175; 135, 5–9.
65 The theme of the unity of human nature is revisited several times using different meta-
phorical images, which in my opinion is evidence of the author’s effort to maintain a
presence sui generis of the particular in the universal. I would like to add to the texts
just quoted a passage where Scotus speaks of a kind of latency of the particular in the
universal. The Master asks if man may have knowledge of himself in his conditions, that
is, in the primordial causes ‘before time’ and in the subsequent things produced in the
course of time; the Disciple answers, ‘in utriusque in a quidem generaliter in causis lat-
enter [which I would render as ‘‘may have knowledge of himself in the primordial causes
as something latent’], in altera vero specialiter in effectibus manifeste. Nam in ila primor-
diali et generali totius humanae naturae conditione [one moment after creation] nemo
se ipsum specialiter cognoscit, neque propriam notitiam sui habere incipit; una enim et
generalis cognitio omnium est ibi, solique deo cognita. Illic nanque omnes homines unus
sunt, ille [The man] profecto ad imaginem dei factus, in quo omnes creati sunt’ (Periphys.
IV , 52, 1423–1429).
66 Periphys. V, 87, 2794–2802: ‘Si humana natura imago et similitudo dei est, profecto et tota
per totum in se ipsa est, et tota in singulis eam partecipantibus, nullam in se ipsa vel in ali-
quo suae uniformiter simplicitatis divisionem vel partitionem seu possibilitate dividendi
vel partiendi seu actu et opere recipiens. Si enim una et individua divinitas est, ad cuius
imaginem humanitas facta est, necessario et ipsa una et individua est, et omnes homines,
nemine excepto, in ea unum sunt.’ See also the analogue text at V, 115.
The Metaphysical System of Scotus Eriugena 85
Where the parts are not at the same time, nor contained in their own
form, the whole cannot exist in a single thing in act or by nature, although
to the attention of the mind the natures of things seem to be at the same
time. But it is different being at the same time by a rational consideration
of nature (which collects all things through the intellect and understands
the inseparable universality) and being a thing that exhausts itself in the
separation or collection of physical parts, through the operation of an
agent, or passion of a patient.67
This might suggest that the unity of the natures of things is a product of the
mind. Soon after, however, Scotus states, with regard to the ousia – though the
same seems to be also true for other universals – that although it is divided by
reason in species and numbers, by its natural virtue it remains individual. It
would seem, therefore, that the nature of the things that leads them to suffer-
ing ‘by a certain fragility’ cannot determine a division of general essences pres-
ent in their individuality to the attention of the mind. Another text that can be
considered is the following (from the Lib. IV):
67
Periphys. I, 70, 2139–2146: ‘Ubi enim partes simul non sunt, neque propria forma conti-
nentur, totum simul esse non potest in aliqua re actu et opere, quamvis simul videantur
esse naturas rerum considerantis cogitattione. Sed aliud est simul esse rationabili naturae
consideratione, quae omnia simul semper intellectu colligit et inseparabiliter compre-
hendit universalitatem, aliud quod operatione agentis aut passione patientis visibiliter in
separatione seu collectione sensibilium partium peragitur.’
86 chapter 3
their notions are naturally in men, it follows that these have been created
universally in men.68
Physical species originate in relation to the essence, together with
quantities, qualities, places, times and other things, without which the
essence would be unintelligible […] the understanding of all things that
the Father created in His Word, is the essence of those things and any
object which is understood in relation to that essence.69
Taken literally, these texts may be slightly embarrassing, since they make
Scotus look like an idealist rather than an extreme realist.70 The notions of the
objects are presented as something which God has placed within the human
soul, and as more real than the objects themselves. However, one must recog-
nize not only that these texts are metaphorical, but also that they are isolated
instances that appear in psycho-epistemic contexts. As the reader may remem-
ber, the soul’s motions have their proper object, which is at least partially inac-
cessible through our cognitive capabilities. Scotus recognizes that the object
in the soul, or of the soul, is more important than the external object, but only
because the mind represents it in a state closer to its essential being. Thus, I
ultimately believe that the views expressed in the texts in which Scotus speaks
of essences and other abstract entities is predominantly ‘externalist’; namely,
what we perceive or think is other than, and external to, the faculty that per-
ceives or thinks it.
68 IV, 48, 1286–1301: ‘Proinde non irrationabiliter iubemus credere et intelligere omnem visi-
bilem et invisibilem creaturam in solo homine esse conditam […] Omnis autem rationa-
bilis natura omni irrationabili et sensibili naturae recta ratione praeponitur […] Si res
ipae in notionibus suis verius quam in se ipsis subsistunt, notitae autem earum homini
naturaliter insunt, in homine igitur universaliter creatae sunt.’
69 IV, 55, 1522–1529: ‘constituuntur autem circa essentias quidem species sensibiles, quan-
titates, qualitates, loca, tempora, et similia, sine quibus essentia intelligi esse non potest
[…] intellectus omnium, quae pater fecit in suo verbo unigenito, essentia eorum est et
cunctorum quae circa eam naturaliter intelliguntur.’
70 In fact, some studies have emphasized an apparent proto-idealism in Scotus; cf.
D. Moran, The Philosophy, esp. 123 et seq. D. Moran, ‘Idealism in Medieval Philosophy: the
Case of Johannes Scottus Eriugena’, Medieval Philosophy and Theology, 8 (1999), 53–82;
W. Beierwaltes, Eriugena. Grundzüge seines Denkens, Klostermann, Frankfurt am Main
1994); For insight into Scotus’ influence on German idealism, see W. Beierwaltes, ‘The
Revaluation of John Scottus Eriugena in German Idealism,’ in J. O’Meara – L. Bieler (eds)
The Mind of Eriugena, Irish University Press, Dublin 1973, p. 190–199.
chapter 4
a. For each proper noun and description there are things corresponding
to it
b. For each predicate expression there is a corresponding property.
c. Properties are sortally different from what corresponds to proper nouns.
d. To say that something denoted by a proper noun has a property means
that the thing and property are the only constituents of a fact (sentence
meaning).
I believe there is a thin line between the explanatory perspective that deter-
mines (and is determined by) a certain metaphysical approach, and the se-
mantic perspective which by itself does not involve that option, but at the
most can be determined or preceded by some form of metaphysical penchant.
Simply stating that ‘Socrates is a man’ means that Socrates has the property of
being a man; as Abelard remarks, we do not have an explanation but a refor-
mulation of what the first statement says. On the other hand, looking towards
modern semantics, for example the standard interpretation of first order pred-
icate logic, the explanatory perspective seems to be less important, as it is dif-
ficult to argue that what justifies the truth of a proposition consists in applying
a function to an argument or coincides with a truth value.
With sets we find ourselves in another borderline situation: obviously set
theory cannot, by itself, provide an explanation of ordinary predication. For in-
stance, the fact that Socrates belongs to the set of men, or a certain set of quali-
fied essences, can hardly explain why Socrates is a man. On the other hand, if we
The designation ‘things that are’ in the table below is taken from the De
Institutione Arithmetica by Boethius. Indeed, the debate on the metaphysics of
predication does not always follow Aristotle’s categorical framework. The uni-
versals at stake are Aristotelian substances, but also the sort of essences that
Boethius describes in De Institutione Arithmetica. Forms that are sometimes
considered for the role of universals are not distinct as elements of different
categories, but rather are intended – as in Priscian – as meanings of predica-
tive expressions (nouns, adjectives and verbs).
(x) (x)
The last two columns have been left blank because philosophers’ positions on
the one-many distinction vary. The idea is that anyone who believes that an
expression in column 2 names a certain entity simply checks the box labelled
‘one’, while one who believes that we can only speak of individual instances
under that expression checks the box labelled ‘many’, and one who believes
that there are types as well as tokens selects both boxes.
A realist is usually expected to be more willing to accept any kind of en-
tity. Thus, he is more inclined to check off all the boxes. Conversely, a nomi-
nalist is generally drawn to desolate landscapes, and thus the empty columns
90 chapter 4
above may indicate a nominalist (actually a very strong one). It seems to me,
however, that there are several variations and it is not easy to know when one
or more checkmarks determine the acceptance or abandonment of a certain
philosophical position. Anyone would identify as a nominalist – without hav-
ing any specific reason – a philosopher who claims there are only individuals
of the type of first substances (as for instance Socrates, Plato, Fido …). Things
that are not located in space and time do not exist. Such a philosopher prob-
ably does not exist in the period we are considering, but if he did exist (say,
for example, somebody called Roscelin) we could set a minimal ontological
standard, and claim that the more a position diverges from that standard, the
more demanding his ontology is.
Realism is a matter of degrees. Especially in the period we are considering,
the boundary between nominalism and realism might seem like a boundary
within the the territory of realism. Rather than invent labels according to the
ontological choices and nuances of metaphysical theories, I consider it more
convenient to identify the extremes of a spectrum. One is the position, per-
haps maintained by Roscelin, that there are only first substances and particu-
lar accidents; the other extreme is the position of William of Champeaux and
his school, according to which particular substances share (or participate in)
second substances and general properties. The following classification helps to
summarize the situation:
nominalism
Realism
a Thus, I reconstruct Walter’s position following a circumstantial paradigm: his theory does
not allow second substances but requires statuses. Likewise, he does not claim – as Abelard
does – that general properties are theoretical entities that do not exist.
As can be seen, the labels ‘realism’ and ‘nominalism’ identify either end of the
spectrum, while the metaphysical theories encompassed by this span can be
judged according to different standards as closer to one or the other extreme.
and at the same time in different individuals, which, being one and not dif-
fering in essence, are multiplied only by the diversity of the accidents.’1 In the
Logica Ingredientibus, Abelard describes this position in the following terms:
Some people consider the universal thing as if the same substance were
essentially in things different by their forms, it being the material essence
of the singular things in which it is, unique in itself, different for lower
forms only. If forms could be separated, there would be no difference be-
tween things that differ from each other only in forms, their matter being
essentially the same. In other words, the substance of man is identical in
numerically different individual men, because of these accidents it be-
comes Socrates, and because of others Plato.2
The first realist thesis thus claims that ‘the same substance is essentially in
things different in form.’ Thesis TR1 involves the following points:
TR1
1. The universal is a unique thing
2. it constitutes the essence of several singular things
3. the differences in the essence are due to ‘inferior’ forms ( formae inferiorum).
4. If one subtracts accidental features, things are indistinguishable
After having presented the three doctrines on universals (with their sup-
porters), we begin to examine the first, the realist one. In this regard, there
are different opinions, as different people say different things about uni-
versals. Some argue that the ten things are different in nature according
Thus, the realists argue that universal things are essentially in the things of
which they are predicated. The clause ‘eadem quae est in Socrate per adueni-
entes formas affecta est Socrates’ does not mean that the animal in Socrates
is Socrates, but that it is Socrates when Socrates’ characteristic properties are
added.
The exposition of TR1 in the De Generali et speciali statu rerum universalium
is similar to Abelard’s:
On the other hand, there exists an ancient doctrinal position, almost fall-
en into oblivion due to ancient faults, that by nature any genus precedes
its inferiors, some forms supervene upon it, reducing the same general
nature to lower things. As we can see in the genus /animal/, prior by na-
ture, in which the physical differences /rational/, /irrational/, /mortal/
and /immortal/ supervene, so that they divide /animal/ and specify the
same undivided genus.4
3 Logica ‘Nostrorum petitioni sociorum’ (= Log. Nostr.), ed B. Geyer, Aschendorff, Münster 1933
(BGPhM, XXI, Heft 4), 515, 10–24: ‘Nunc autem tribus sententiis de uniuersalibus positis cum
testimoniis ipsarum, primam quae de rebus est, primitus exsequamur. De qua etiam sunt
plures opiniones, cum alii aliter res uniuersales esse affirmant. Nonnulli enim ponunt decem
res diuersas esse naturaliter secundum decem praedicamentorum uel generalissimorum dis-
tinctionem, cum uidelicet ita dicant res esse uniuersales, hoc est naturaliter communicabiles
pluribus, quod eandem rem essentialiter in pluribus ita ponunt ut eadem quae est in hac re,
essentialiter sit in illa, diuersis tamen formis affecta. Verbi gratia ut animal, natura scilicet
substantia animata sensibilis, ita est in Socrate et Brunello et in aliis, quod eadem quae est in
Socrate et per aduenientes formas effecta est Socrates, et essentialiter tota est in Brunello ita,
quod Socrates nullo modo a Brunello in essentia diuersus est sed in formis.’
4 De Generali et speciali statu rerum universalium (= De Gen. Stat.), ed. Dijs (in J. Dijs, ‘Two
Anonymous 12th-Century Tracts on Universals’, Vivarium, 28, 1990), 93–94; ed. Romano (in
Una soluzione originale della questione degli universali nel XII secolo, Aracne, Roma 2007),
p. 91: ‘Est autem antiqua sententia et quasi antiquis erroribus inveterata quod unumquodque
94 chapter 4
The genus here is ‘naturally prior’ to inferior things (in the order relation of
predication). Walter and Abelard use the verb ‘to supervene’5 to characterize
the manner in which forms apply to a common nature. The animal, for ex-
ample, is considered to be a natural res which precedes its inferiors (or ‘lies
before’ them = preiacet) waiting for properties. If we remove from the animal
the forms and properties by which it is divided up, we obtain the universal
animal once again. Apparently Walter of Mortagne (supposed author of the
De Gen. Stat.)6 has a rather special case in mind: accidental properties cannot
lead to a specification of the genus; on the other hand, the text quoted speaks
in general of forms (essential and accidental) that can also be applied to the
genus to form either the species or an individual. The addition of forms is, at
least conceptually, a reversible operation,7 since subtracting forms should take
us back to the common nature.
The presentation in the De Generibus et speciebus is more hasty.8 According
to the author, probably Joscelin, there were two parties among realists.
According to the first, individuals are identical to genera and species, for those
of the second party genera and species are common essences present in indi-
viduals. Both positions are labeled as realist, as opposed to those according to
which species and genera are linguistic expressions. Strictly speaking, only the
second position can be classified as realist, since for identity theorists there
are no universal entities. The main thesis of realism (TR1) is summarized in the
following terms:
genus naturaliter preiacet suis inferioribus, cui naturaliter preiacenti superveniunt formae
quaedam, quae redigunt ipsam generalem naturam ad inferiora. Sicut in animali genere vi-
dere possumus quod in natura preiacet, cui superveniunt hae differentiae, rationale et irra-
zionale, mortale et immortale, quae animal dividunt et ipsum divisum specificant.’
5 The notion of supervenience is often cited in 12th century literature on universals. It is still
used today in some areas of philosophical inquiry (cf. the article by B. McLaughlin and
K. Bennett in the Stanford Encyclopedia of Philosophy), and seems to retain a certain charm,
which I believe is due to the air of mystery and obscurity surrounding it.
6 Cf. paragraph 6.1 infra.
7 Walter considers the case in which properties individualizing the animal are destroyed;
should, for example, the properties of being rational or irrational be destroyed, there would
no longer be any individuals, nor second substances, following the Aristotelian thesis accord-
ing to which second substances cease to exist as well when first substances are removed.
8 De Generibus et speciebus (= De Gen.), in P. King, ‘Pseudo-Joscelin: Treatise on Genera and
Species’, Oxford Studies in Medieval Philosophy, 2 (2014) 132–211.
Realist Theories in the 11th–12th Centuries 95
/Man/ is a certain species, essentially one thing to which forms are ad-
joined which give rise to Socrates; other forms shape the same essence
constituting Plato and the remaining individuals of the human species.9
In Socrates – as reported by the author – there is nothing other than the spe-
cific matter and individuating forms. The same matter is identical in Plato and
in other men. Just like species, or specific material, is located in the individuals,
in the same way the genus matter is located in the species.
TR1 in some way reflects the traditional structure of reality suggested by the
Porphyrian tree; we can distinguish between universal substances, essential
forms, forms not constitutive of species and the individuals under them:
/substance/
/substance/ + /form of living/ = /living being/
/substance/ + /form of living/ + /form of rationality/ = /man/
/substance/ + /form of living/ + /form of rationality/ + individual forms
= Socrates
Apart from the highest genus which does not contain forms, all subordinate
genera are made in such a way as to be what they are through forms adjoined
to a higher genus. When different forms are added the universal is divided into
species, and it returns to being one by removing the forms which were added.
From the point of view of TR1 the universal is a constituent element of in-
dividual things. Just as in the classic metaphor of the statue, the same block of
wax can be molded into different shapes, in the same way the universal gives
rise to different individuals. Abelard notes that this metaphor is limited insofar
as the same wax cannot simultaneously constitute different statues, while the
universal is by definition a totality that, being common to different things at
the same time, ‘constitutes’ their substance ‘materially’. Such a totality in itself
is universal and becomes singular due to the forms that are added to it.
9 De Gen., ed. cit., 33 (134), 7–10: ‘Homo quaedam species est, res una essentialiter cui adveni-
unt formae quaedam et efficiunt Socratem. Illam eamdem essentialiter eodem modo nunc
informant formae facientes Platonem et caetera individua hominis.’
96 chapter 4
[Boethius believes that] the same [universal] is particular for the ad-
joined forms, without which it naturally subsists in itself, but it does not
remain in act regardless of the same forms; so it is universal by nature and
particular in act, and it is understood as incorporeal and unperceptible in
the simplicity of its universality, while it subsists in act as perceptible and
corporeal for the accidents.10
This is not exactly what Boethius says, however. Rather, his version of the Neo-
Platonic theory on the different ways of being of the universal is:
Genera and species subsist in a certain way and are thought otherwise,
and they are incorporeal, yet connected with bodies they exist in par-
ticular things, they are understood as subsisting in themselves and not as
having their being in something else.11
It should be carefully noted that the thesis of universality ‘by nature’ is attribut-
ed to Boethius, and not to some contemporary realist. Later in the text Abelard
states that the discord between Aristotle and Plato lies in disagreement about
the thesis of the actual presence of the universal in the particulars. Abelard
obviously knows nothing about the metaphysics of either Plato or Aristotle;
his Plato claims that the specific/general nature subsists in itself, but is then far
more modest and conciliatory about a natural ability to subsist. What matters,
ultimately, is how Abelard utilizes the (alleged) opinions of the authorities to
introduce his criticism to the realist position.
In presenting the realist thesis, Walter considers the problem of how the
genus can remain once the differences have been removed, i.e. whether the
addition of the differences is reversible. Walter excludes that supporters of TR1
can accept the permanence in act of the genus. The argument runs as follows:
if we destroy the properties that allow us to specify the genus, we destroy the
individual; but according to Aristotle, without individuals there are no species
and genera, and therefore genera do not remain in act once the differences are
destroyed, though they can potentially remain. It is not entirely clear whether
10
Log. Ing., 11, 4–7: ‘Idem per advenientes formas singulare sit, sine quibus naturaliter in
se subsistit et absque eis nullatenus actualiter permanet. Universale quidem in natura,
singulare vero actu et incorporeum quidem et inseparabile in simplicitate universalitatis
suae intelligitur, corporeum vero atque sensibile idem per accidentia in actu subsistit.’
11
In Isag., 167, 8–12: ‘Ipsa enim genera et species subsistunt quidem alio modo, intelliguntur
vero alio, et sunt incorporalia, sed sensibilibus iuncta subsistunt in sensibilibus, intel-
liguntur vero ut per semet ipsa subsistentia ac non in aliis esse suum habentia.’
Realist Theories in the 11th–12th Centuries 97
1. The expressions ‘act’ and ‘potency’ are used as parts of verb modifiers in phras-
es such as ‘actually laughs’ or ‘has the potency to laugh’ (‘can laugh’). ‘Tantum
interest’ Boethius states ‘actus a potestate, quantum homo ridens ab eo qui ridere
possit, not tamen rideat.12 In the Dialectic Abelard points out that something
which is not currently white has the possibility of being white, and he adds
that when one speaks of potency in this sense a natural ‘non-repugnance’
is at stake. This meaning of ‘potency’ is halfway between logical possibility and
the capacity to be in accordance with the current state of the world.
2. Accidents are usually13 considered separable by nature or separable in
act. An example of the first type is the property of being bipedal, of the second
type being seated. As such, the accidental property of being bipedal is not nec-
essarily inherent in man, and thus it is separable. On the other hand, no man is
currently not bipedal, and thus the same property is not separable ‘in act’. One
may think that the distinction has to do with the Divine plan: men could have
been different, but according to the way the world was created, they are as they
are. A dual conception of modalities is present here: in one sense, accidents
are not necessarily inherent in a subject, therefore we cannot say that man is
necessarily bipedal, and in another sense it is necessary what is always in act,
or, as Abelard says, that whose act is eternal. Both concepts are to be distinct
from the ‘logical’ or ‘synchronic’ concept of possibilities.14
The same opponents of realism recognized that the observations on the pres-
ence in act/potency of forms in individuals are convincing and based on au-
thoritative sources. On the other hand, they do not address a more acceptable
variant of TR1, and criticize only the main thesis. Abelard makes the distinc-
tion in order to explain the positions of Plato, Aristotle and Boethius, and
Walter apparently refers to it while suggesting a lenient interpretation of TR1.
In these criticisms of realism, we do not find counterarguments exploiting the
distinction between act and potency. Thus, a variant of TR1 does not appear to
be hidden behind Walter’s and Abelard’s considerations.
It still remains to be discovered what kind of realism it is that considers
potential universals. The arguments used to counteract the thesis of the pres-
ence of contrary properties in the genus (see the discussion of this aspect in
For some observations on Abelard’s position, cf. R. Pinzani, The Logical Grammar of
Abelard, Kluwer, Dordrecht 2003.
15 This position depends on Porphyry’s claim that ‘The difference is what the species has in
addition to the genus … [genus] certainly has differences in potency under it, but none in
act.’
Realist Theories in the 11th–12th Centuries 99
the following chapters) show how difficult it is to combine the TR1 with the
doctrine expounded by Porphyry in the Isagoge. It is likely that, by taking seri-
ously Abelard’s distinction between subsistence by nature and presence in act,
one does not preserve TR1 but moves towards TR3.
There are others who – in respect of this old thesis [TR1] – do not admit
that the same thing is present in Socrates and Plato, because, since man is
a natural thing per se, received socrates-ness, he turns out to be nothing
other than Socrates, and received plato-ness, he turns out to be nothing
other than Plato. However, they concede that the same natural entity is
in different things, and not personally identical; it is different therefore to
speak of the simple man, of man in relation to inferior things or of man
as made inferior.18
16 Historia calamitatum, Ed. cit, 65, 89–91: ‘Sic autem istam tunc suam correxit sententiam,
ut deinceps rem eamdem non essentialiter sed indifferenter diceret.’
17 I translate here the Latin expressions ‘indifferentia’, ‘indifferens’ as ‘non-difference’ and
‘non-different’.
18 De Gen. Stat., ed. Dijs 94; ed. Romano, 92: ‘Sunt autem alii qui non concedunt in hac sen-
tentia antiqua eamdem rem actualem esse in Socrate et in Platone, quia, cum homo sit
res naturalis per se, suscepta socratitate, non est aliud homo quam Socrates, et suscepta
platonitate non est aliud quam Plato; se tamen concedunt eamdem rem naturalem esse
in diversis, non eamdem personaliter; et est aliud agere de homine simplici, aliud de ho-
mine circa inferiora, aliud de homine inferiorato.’
100 chapter 4
19 Re-interpreted in the Questiones (i.e. the text of BN 17813 of the National Library of Paris,
published by Hareaux, so called from its opening words).
20 De Gen. Stat., ed. Dijs 96, 7; ed. Romano, 93, 8: ‘mihi autem videtur quod quicumque huic
sententiae consentiunt, veritati contradicunt dum fabulam tenent’.
Realist Theories in the 11th–12th Centuries 101
With respect to TR3, Walter observes that man can be considered ‘in his
simplicity’, regardless of individual instances. For example, when we claim
‘man is a species’ (homo est species), the subject of the speech is the simple
man (homo simplex) and by no means particular men. This way of putting
things is rather surprising: although other authors recognize that ‘man’ in ‘man
is a species’ has a special meaning (Abelard speaks of a shifted meaning), the
terminology and the solution proposed by Walter suggest logical analyses de-
veloped by the logicians of 13th and 14th centuries. Walter continues his expo-
sition of TR3, stating that when we refer to the single men, we consider man
‘in the inferiors’ and no longer man in his simplicity. This explanation is rather
unexpected, as it is not metaphysical but logical:
One who holds TR3 apparently does not accept that the simple man can be just
the same as Socrates, and also that a simple sentence that concerns the man
can be phrased in terms of a quantified sentence. In support of this position,
Walter invokes the authority of Boethius, Priscian and Porphyry. His mention
21
De Gen. Stat., Ed. Dijs, 95; ed. Romano, 92: ‘Cum vero agimus de homine circa inferiora,
non simplicitatem hominis attendimus, sed hominem in inferioribus, ut cum ‘homo est
animal’ dicimus, hominem circa inferiora subicimus.’
102 chapter 4
Considering that in any case a universal There are others who deem the
that concerns things must be admitted, universality a thing, and consider the
they say that particular things are identical same entity universal and particular. In
(idem), not essentially but by non- fact they assert that the same thing is
difference […] In other words, individual found in several objects inferior by non-
men, particular in themselves, are the same difference and not essentially. So they
thing in the man, that is they do not differ say the same thing (idem) is in Socrates
in human nature […] and the same things and Plato; by ‘the same thing’ they
that are called particular according to their mean ‘that which is non-different’ or
22 The identification of the supporter of TR3 with William’s student is only a suggestion,
since there is no conclusive evidence for thinking that the originator of TR3 is the
Abelard’s enemy, or rather one of the many enemies from William’s school.
23 Log. Ing., 13, 18–25: ‘Unde alii aliter de uniuersalitate rerum sentientes magisque ad sen-
tentiam rei accedentes dicunt res singulas non solum formis ab inuicem esse diuersas,
uerum personaliter in suis essentiis esse discretas nec ullo modo id quod in una est, esse
in alia, siue illud materia sit siue forma, nec eas formis quoque remotis minus in essentiis
suis discretas posse subsistere, quia earum discretio personalis, secundum quam scilicet
haec non est illa, non per formas fit sed est per ipsam essentiae diuersitatem.’
Realist Theories in the 11th–12th Centuries 103
particularity, are called universal according ‘similar’. When they say that the same
to non-difference and similarity.a thing is predicated of several things or
is inherent in something, they mean
that several things are alike in some
nature, that is, are similar. For example,
things are alike in being bodies or
animals. According to this doctrine,
as we have said, the same thing is
universal and particular, considered
however from different points of view
[…] According to those people ‘to be
predicated of several things’ means
‘there is a certain status in which
several things participate.b
a Log. Ing., 14, 1–6: ‘uniuersale tamen rerum adhuc retinentes idem non essentialiter quidem
sed indifferenter ea quae discreta sunt, appellant, ueluti singulos homines in se ipsis discre-
tos idem esse in homine dicunt, id est non differre in natura humanitatis, et eosdem quos
singulares dicunt secundum discretionem, uniuersales dicunt secundum indifferentiam et
similitudinis conuenientiam.’
b Log. Nostr., 518, 9–26: ‘Sunt alii in rebus uniuersalitatem assignantes, qui eandem rem uniuer-
salem et particularem esse astruunt. Hi namque eandem rem in diuersis indifferenter non
essentialiter esse inferioribus affirma<n>t. Veluti cum dicunt idem esse in Socrate et Platone,
‘idem’ pro indifferenti, id est consimili, intelligunt. Et cum dicunt idem de pluribus praedi-
cari uel inesse aliquibus, tale est ac si aperte diceretur: quaedam in aliqua conuenire natura,
id est similia esse, ut in eo quod corpora sunt uel animalia. Et iuxta hanc, ut diximus, senten-
tiam eandem rem uniuersalem et particularem esse concedunt, diuersis tamen respectibus
[…] tale secundum illos ‘praedicari de pluribus’ ac si dicatur: Aliquis status est, participa-
tione cuius multae sunt conuenientes.’
Prima facie, the thesis at issue is quite simple: the same things are universal
or particular considered from different points of view. This thesis is close to
or coincident with that of the Questiones and De Gen. Statu. However, there
are points in the two texts which are unclear. In the Logica Ingredientibus sin-
gle men are one thing in the man, i.e. they are no different in human nature;
apparently on the one hand we have individual things, on the other /man/
104 chapter 4
or human nature, against the thesis of the identity of individual things and
human nature. In the text of the Nostrorum (on the right of the table) it is said
that a certain identical thing is in particular things, again in contrast with the
identity thesis. In both cases the contrast disappears in the transition from TR2
to TR3.
There are two ways of reading the above texts. One can assume that there
are only individual things and explains their ‘universality’ or their ‘being in
other things’ by saying that they are in a certain relation (similarity, identity,
or something else). This perspective has little to do with the kind of realism
that we are considering here, and would be more compatible with some form
of nominalism. From another point of view, however, one can detect a genu-
ine realist position (in the same vein as TR3) that brings together individual
substances and status (or universal properties). In the case of the former posi-
tion, the relation of predication between individual things is defined as a kind
of spurious identity, and in the case of the latter the same relation is defined
between ‘similar’ things and status (common natures).24 The exposition of this
point in the Historia Calamitatum and in the Ingredientibus reveal some gaps;
Abelard tells us that there are individual essences in individual things. I do
not think it is the case that Abelard has insufficient information, nor that he is
reticent on this question; on the contrary, what he says in the following quote
contains, in my opinion, the key to interpreting TR2:
On this point there is some disagreement. In fact, some believe that the
universal thing is nothing but a collection of many things: they do not call
Socrates and Plato species at all, but say the human species is made up of
all men gathered together. […] According to other people the noun ‘spe-
cies’ not only refers to men gathered together, but also to the individuals
in that they are men.25
24 In the Ingredientibus the term ‘status’ does not occur in the context of the discussion of
the theory of non-difference (‘indifference’), but we find expressions such as ‘in homine
convenire’, ‘non differt Socrates a Platone in homine’. ‘Status’ is a technical term that refers
to the meaning of a common noun, for example, the noun ‘man’ means the status of man
i.e. being a man.
25 Log. Ing., 14, 7–19: ‘Sed hic quoque dissensio est. Nam quidam universalem rem non nisi
in collectione plurium sumunt; qui Socratem et Platonem per se nullo modo speciem
vocant, sed omnes homines simul collectos […] Alii vero sunt qui non solum collectos
homines speciem dicunt, verum etiam singulos in eo quod homines sunt.’
Realist Theories in the 11th–12th Centuries 105
The two theses derived from TR2 are very similar to, if not coincident with, the
positions of Walter and Joscelin (or Goslenus, alleged author of De Generibus
et Speciebus). Curiously, Abelard connects TR2 to the positions of authors
who would doubtless be very much disagree with such a relationship. For
Abelard these theories are, naturally, equally as objectionable as William of
Champeaux’s initial position, especially since they speak of things that are
predicated. However, there may be reason for not putting together in one bas-
ket all those who think there are predicated things; in the table below one can
see the different realist positions (‘=‘, ‘≠’, ‘↑’, ‘≈’ mean , respectively , ‘personally
(essentially) identical to’, ‘personally (essentially) different from’, ‘present in’,
‘not unlike’):
Socrates ≠ Plato
↑ ↑
Essence of man = Essence of man
Socrates ≠ Plato
↑ ↑
Essence of man in Essence of man in
≈
Socrates Plato
TR3
Socrates ≠ Plato
|| ||
Man in Socrates Man in Plato
≠
(+socrates-ness) (+plato-ness)
↑ ↑
… …
Essence of man = Essence of man
106 chapter 4
Walter’s Thesis
Socrates ≠ Plato
|| ||
Socrates qua man ≈in being human Plato qua man
The ‘official story’ on the issue of universals begins with Porphyry’s Isagoge,
translated and interpreted by Boethius, and continues with the commentaries
on Boethius’ commentaries, as well as with independent sections in different
works devoted to the discussion of this topic. The authors we have considered,
with the exception of John of Salisbury, share a certain rigid acceptance of the
following points:
Thesis (c) is subject to various interpretations: for example, authors who are
not able to provide a convincing explanation of the real presence of the species
in the genus, but nevertheless want to preserve this point, can try to defend
some metaphorical reading.
Realist Theories in the 11th–12th Centuries 107
The authors in the third column, of course, are quoted as sources and not as
followers of the thesis in the first column. What is left out or remains on the
margins of the ‘official story’ are the properties. However, there is another
‘unofficial’ story, which can be outlined as follows:
26 John of Salisbury, Metalogicon, ed. J.B. Hall, Brepols, Turnhout 1991, 83, 104–107 (PL 876 A).
27 Translated here either as ‘similarity’ or ‘likeness’, dependent upon context.
Realist Theories in the 11th–12th Centuries 109
the common likeness.’30 The text may suggest an overlap between the notion
of status and that of similarity. This would suggest an intriguing connection
between the Boethian and Abelardian solutions. I believe, however, that the
two concepts should not be overlapped, as likenesses are mental and their real
counterparts resemble relational properties, whereas statuses are non-mental
and look like non-relational properties. The formulation ‘conceived likeness’
remains ambiguous: does it refer to a mental image represented in the intel-
lect, or to a real likeness which the intellect thinks about?
e) Abelard believes that words such as ‘equal’, ‘like’, and ‘contrary’ refer to
relations. Similarity in particular is defined as a kind of relationship which
‘contains’ different individuals, as these are considered in relation to each
other (sese respiciunt). The adjective ‘similar’, which derives (sumptum est)
from ‘similarity’, is considered a relative according to signification as well as
construction (in the sense that it contributes to form a relational phrase).
Abelard gives the example of a body that is similar to another because of its
whiteness; it does not possess the same whiteness, but whiteness is that by
which the two bodies are similar.
Why do authors not talk about similarity where they should, i.e. in the context
of the commentary (on the Boethian commentary) on Porphyry? I believe this
reluctance to discuss the issue of similarity is a result of the aforementioned
rigidity in posing the problem of universals. The authors are aware of the im-
portance of the concept of similarity, yet seem to have difficulty understanding
how a universal entity – if any – may differ categorically from its Urelementen:
relations, as well as qualities, or perhaps even more than qualities, cannot be
universal with respect to first substances, but only to particular qualities or
relations.
Points 1–6 above outline an ‘underground history’ of the problem: to some
extent we are outside the traditional framework, but this does not mean, at
least in principle, that once universal properties are introduced one ought
to stop speaking of the species or genera, or of the relation of predication.
Properties appear several times in this work in different contexts and in vari-
ous forms. In 2.2 I considered Boethius’ logical approach according to which a
categorical statement means the inherence of a property (form) in a subject;
here I would like to anticipate some observations on the theory of predication
in Abelard, to show how the issue of the inherence of a property in a subject is
dealt with independently of that of the universals (except for an interference
I consider below).
30
Log. Ing., 19, 9–13.
Realist Theories in the 11th–12th Centuries 111
[After the general issue of the inherence of the predicate], there remains
the question of the four inherence modes of the same predicate […] It
is necessary that the predicate is considered as a linguistic expression,
[…] In fact, the definition is not a predicate belonging to the real world,
nor is the genus, since according to our doctrine it is a mere linguistic
expression, as we concede that only these linguistic expressions are uni-
versal, i.e. predicable of several things. But perhaps those who also con-
sider things universal, when [Boethius] says that the genus, accident and
propia are things, would designate by the noun ‘definition’, by a transfer
of meaning, the thing signified by the definition.31
31 Abelard, Super Topica Glossae (= SupTop.), ed. M. Dal Pra, in Pietro Abelardo, Scritti di
Logica, La Nuova Italia, Firenze 1969, 270, 30–271, 5: ‘Restat questio quisnam modus
de quattuor inhaerendi […] Oportet praedicatum accipi vocale […] quippe reale
112 chapter 4
Here Abelard sides with those who consider only linguistic expressions to be
universal, in order to show how the problem of interpretation is more easily
praedicatum diffinitio non est, nec etiam genus secundum nostram sententiam dicitur
nisi vox, quia solas voces universales, id est de multis praedicabiles concedimus esse. At
fortasse hii qui res quoque universales esse concedunt, cum genus aut proprium vel acci-
dens res esse dixit, diffinitionis quoque nomine per quamdam translationem significatam
a diffinitione designabunt.’
32
SupTop., 274, 20–275, 10: ‘ubi socrates dicitur esse ipsum subiectum albedinis, simul os-
tenditur ipse albedine adiacenter affici et duo socrati copulari videntur: albedo quidam
in adiacentia, subiectum albedinis in essenstia; solum tamen albedo dicitur praedicari,
licet non solum copuletur, quia si rerum praedicationem in universalibus quoque vocibus
sicut alii attendimus, non oportet vocari rem praedicativam esse illam quae copuletur
subiecto, sed illam tantum quam propositio copulare intendit et ad quam copulandam
praedicatum vocabulum ponitur, sicut album hoc nomen de socrate enuntiatum maxime
in significatione albedinis coniungendae subiecto ponitur […] Quod autem dicitur supra
[…] utrum praedicatum inhereat subiecto, de rerum sensu hic non de vocibus accipimus.’
Realist Theories in the 11th–12th Centuries 113
There exist some testimonies (among others Abelard’s) about a twofold way
of reading the texts of the Logica vetus between the eleventh and twelfth cen-
turies: according to some, Aristotle and Porphyry were primarily interested in
discussing linguistic issues, while others claim that the same authors also or
mainly dealt with real things. The panorama of ‘nominal’ commentaries on the
texts of the logica vetus is actually quite varied. The edited texts allow us to get
a fairly precise idea of the way in which authors read Boethius and Porphyry
in voce.2
Some commentaries look like brief introductions or a kind of exercise, in
which brief remarks follow phrases from the text to be commented. An exam-
ple is constituted by the texts edited by Y. Iwakuma and a part of the commen-
tary on the Categories contained in the so-called Introductiones Dialecticae
(= Literal Glosses: cf. below). In these texts authors – labeled by scholars as pre-
or proto-nominal – while arguing that Boethius and Porphyry are concerned
with logical expressions, do not take positions on metaphysical issues such as
that of properties or the status of genera and species. Besides, there are other
(sections of) commentaries in which the authors do not evade the task of ex-
plaining passages with a clear realist flavor to them.
1 Under the title ‘Literal Glosses’ I refer to the commentaries on the Categories and Isagoge
contained in Paris ms lat.13368 edited by M. Dal Pra (P. Abelardo, Scritti di Logica, La Nuova
Italia, Milano 1969). The attribution of these glosses to Abelard has been questioned by
C. Martin (‘A note on the attribution of the Literal Glosses in Paris, BnF lat. 13368 to Peter
Abelard’, in I. Rosier-Catach, Arts du language et théologie aux confins des XIe et XIIe siècles,
Brepols, 2011, pp. 605–646) and M. Cameron, ‘Abelard’s early glosses: some questions’ (Ibid.,
647–662). I will maintain Abelard’s authorship here (see the initial observation about ‘local
regulations’),
2 Y. Iwakuma, ‘‘Vocales’ or early Nominalists,’ Traditio, XLVII, 1992; id., ‘Twelfth-Century
Nominales. The Posthumous School of Peter Abelard’, in Vivarium 1992 (XXX, 1), 97–109.
A collection of studies on the origins of medieval nominalism is contained in the special
issue of Vivarium (1992) XXX–1 (art. by: Mews, Kneepkens, Marenbon, Ebbesen, Normore,
Iwakuma, Martin, Luscombe, Colish, Coutenay).
3 J. Marenbon, ‘Logic at the turn of the twelfth century’, in D.M. Gabbay – J. Wood (eds.),
Handbook of the History of Logic, vol. 2 Medieval and Renaissance Logic, Elsevier, Amsterdam
2008, 65–81, p.77.
4 Whilst it is not too difficult to fix the limits of the interval between realism and nominalism
(on one hand everything we can talk about exists, on the other only individual things that
can be reached in some way by our sensorial apparatus), it does prove complicated to estab-
lish the point in the interval where one of the two philosophical options prevails. We might
wonder, for example, whether Abelard is more a nominalist or a realist; the same applies – in
more recent times – to Russell (who suggests an atomistic ontology with properties), to the
‘scientific’ realists, etc.
5 St. Aselm, Opera Omnia, Vol. II, ed F.S. Schmitt, Fromman Verlag, Stuugart – Bad Cannstatt
1984, 9, 21–22: ‘Illi utique nostri temporis dialectici, immo dialecticae haeretici, qui non nisi
flatum vocis putant universales esse substantias.’
116 Chapter 5
A certain quantity of textual material on the first nominals has been collected
in recent times. However, we have only a few (almost) complete works that we
can safely place within the nominal school. In this chapter I consider two texts
by followers of the nominalis secta, and focus on the authors’ attempts to go
beyond a mere syntactic reading of the Categories and Isagoge.
Based on these texts one can propose an interpretative model, or, more
modestly, a reading hypothesis. One perspective is to view the lectio nominis as
a meta-linguistic interpretation of the principle of predication (see Chapter 1).
Let us consider the following equivalences:
The first part of the Dialectica by Garland the Computist6 contains com-
mentaries on the Isagoge and Aristotle’s Categories. There is no trace of the
6 I will call the author of the Dialectica ‘Garland the Computist’, even though, according
to recent studies, the author is said not be Garland the Computist but a certain Gerland
(from Besançon?). On the author of the Dialectica and the dating of the work, see De Rijk’s
Introduction to the critical edition (Van Gorcum, Assen 1959) and Y. Iwakuma, ‘‘Vocales’ or
early Nominalists,’ cit. The former author proposes an ‘early’ dating from 1040–1070, and the
The ‘ Nominal ’ Stance 117
The text focuses on linguistic predication; Garland’s aim may be to reduce all
threads to grammar problems or build a bridge between real events and facts
of grammar. The text above simply provides a clue, but – unlike Roscelin –
we are in a position to follow the development of the nominal interpretation.
First, though, a word of warning is necessary: Garland’s position is somewhat
unstable, since he fluctuates between expressly declaring his adherence to
the nominal interpretation, and at other times speaking of genera and spe-
cies, using and not mentioning the related terms; at still other points he talks
about the meanings of general terms. Having found no real justification for this
change in perspective, I believe it is better to faithfully follow the text of the
Dialectica than to attempt to ‘tidy up’ the texts according to some theoretical
position, either declared or supposed.
Shortly after explaining that a genus is simply a linguistic expression,
Garland accidentally slips onto the non-linguistic plane in the following
latter a ‘later’ one from 1110–1120/1130. The two theses are argued on the basis of palaeographi-
cal and textual evidence too; I cannot suggest alternative datings here, I just want to draw
attention to the fact that the author of Dialectica (whom I will call Garland the Computist) is
a clever dialectic. If we place the work in the context of the debate on Porphyry and Boethius’
comments, i.e. after 1110, one cannot explain the absence of any reference to this debate and
the different positions which were shaping in that period. At the same time I think it would
be reasonable not to backdate the text of Dialectica too much (as in De Rijk’s ‘very early’
dating).
7 Dialectica, ed. cit., 3, 13–17: ‘Tercium autem, de quo agunt phylosophi, sic diffinitur: genus est
quod predicatur de pluribus differentibus specie in eo quod quid sit; ‘quod predicatur’ facit
differre ab illis que non predicantur, scilicet a rebus, non vocibus, que non habent poni in
propositione.’
118 Chapter 5
example: ‘man is predicated of Socrates and Plato, but Socrates and Plato do
not differ in their species, namely in being human (per hominem).’8 While the
first part of the example suggests that Garland accepts the less radical nominal
principle (‘X’ is predicated of x iff x is an X), the second part presents a use of
the term ‘man’ which is difficult to interpret: it could mean that Socrates and
Plato do not differ in the noun common to them, or that they do not differ in
having a certain real common feature. In the latter case one may wonder what
the limits of Garland’s faith in the lectio nominis really are. In the text that fol-
lows the nominal point of view once again becomes prominent:
So, /substance/, which is the higher expression (vox), is the highest term,
/man/ and God, which are the inferior species, are called particular spe-
cies and /animal/, /body/ and the others which are at an intermediate
level, are called genera subordinates […] ‘body’ is in fact a subaltern
8 Dialectica, 3, 28–4, 2
9 Dialectica, 5, 27–31: ‘Predicamentum autem vocamus unam collectionem vocum quarum una
nascitur sub alia, sicuti substantia et voces sibi supposite. Predicamentum autem ideo voca-
mus istam vocem que est ‘substantia’ cum sibi suppositis vocibus quia genus predicatur ibi
de suis speciebus et omnia superiora de inferioribus.’
The ‘ Nominal ’ Stance 119
A realist interpre-
tation of ‘animal’
/animal/
↓ ↓
→
←
A realist interpre-
tation of /man/ ‘man’
↓ ↓
Plato Socrates Cicero ‘Plato’ ‘Socrates’ ‘Cicero’
It is actually rather surprising that in the text just quoted Garland switches
from a discourse on God to one on the name of God without any comment.
There is only a small amount of vanity detectable (Garland actually defines
himself as a lovable humble person), which was probably common among the
masters of dialectics even before Abelard’s time. Leaving the vanity aside, how-
ever, the problem still remains of figuring out whether the first tree and the
connecting arrows can be justified.
The second chapter is devoted to the species, and provides further infor-
mation on the relationship between the two predication trees. According to
Garland, the term ‘species’ can refer to things as well as to linguistic expres-
sions. Seen as res, the species is the form of anything, since whatever can be
10
Dialectica, 6, 14–20: ‘Ecce hic substantia, que est superior vox, est generalissimum; homo
autem et Deus, que sunt inferiores species, specialissime species appellantur; animal vero
et corpus et cetera que sunt in medio, subalterna genera et subalterne species sunt […]
Vox enim subalterna est ‘corpus’, quia trahit originem ab ipsa substantia, et ‘animatum
corpus’ hec alia vox nascitur sub ea.’
120 Chapter 5
11 Dialectica, 7, 8–9: ‘In rebus appellatur cuiusvis forma species, quia per formam suam qui-
libet inspiciatur’.
12 Garland claims that the species, according to the interpretation that makes it an expres-
sion, is a sermo. I translate ‘sermo’ with ‘meaningful expression’. The technical use of
‘sermo’ depends on that of Boethius’ ‘interpretatio’, considered as an expression to which
a certain conventional meaning is assigned.
The ‘ Nominal ’ Stance 121
Garland recognizes that the difference that philosophers take into consider-
ation is the one on the last line of the table. Here too the language is rather
ambiguous: on the one hand he talks about differences as expressions, on the
other he seems to speak of inseparable real accidents such as blindness. The
following comment, however, is more explicit:
The difference is that expression which stems from (tracta) the genus (as
‘mortal’ stems from ‘animal’) for which the expression consisting of the
genus and the difference itself is wider (habundat) […] that expression
acquired by the genus and which constitutes together with it the species,
always goes together with the species which it constitutes, as for example
the expression ‘mortal’ is acquired by the genus ‘rational animal’ and con-
nected to this genus constitutes the species ‘man’ […] the difference does
not always follow the genus – as it is not true that every rational animal
is mortal.13
The task of the translator is simplified in this case because the use of quota-
tion marks is justified by the context. However, there are some problems. Let
us start from the word ‘trahitur’, which I have translated as ‘being acquired’.
Alternatively, ‘tracta’ could mean something like ‘extracted’ or ‘gathered’, in-
tending the operation of obtaining the species by ‘extracting’ or ‘activating’ an
inner property present in the genus. However, in this latter case it is difficult to
think of a property that is present in the genus in potency, since we are deal-
ing with linguistic expressions. Similar considerations apply to the concept
of abounding (habundare). One may be led to believe that Garland interprets
categorical terms in an intensional way, but in this case, again, they would be
properties or formes, and not linguistic expressions. The ‘abounding’ must
therefore have something to do with the presence of an expression in another
expression: the string ‘mortal rational animal’ is longer than ‘rational animal’
because it contains the expression ‘mortal’ as well as all expressions contained
in ‘rational animal’.
13
Dialectica, 8, 18–28: ‘Differentia est illud, id est illa vox qua voce tracta a genere, sicuti
mortale trahitur ab animali, species constituta ab ipso genere et ab ipsa voce que trahitur
ab illo, habundat […] vox illa que trahitur a genere et que constituit cum illo speciem,
semper sequitur illam speciem quam constituit, ut hic patet: mortale ista vox trahitur ab
isto genere quod est animal rationale, et iuncta cum illo constituit illam speciem que est
homo […] differentia genus non semper sequitur, quia non verum est esse mortale omne
animal rationale’
122 Chapter 5
The last observation in the passage above: ‘the difference does not always
follow the genus – as it is not true that every rational animal is mortal’ estab-
lishes a link between linguistic relations and categorical sentences. The dif-
ference and genus are not in a relation of predication because the categorical
sentence composed by the same terms is not true. We do not have a version
of the principle of predication, but something similar that can be used to jus-
tify a nominal interpretation of Porphyry’s text. The positive side is that one
understands why Garland tries to map real relations onto linguistic relations
(redefining traditional logical notions and introducing new ones) and how
he can accomplish this; the negative side is that the nominal reading of the
Aristotelian principle cannot be used to solve Porphyry’s problem, but at the
most suggests a way to avoid (or postpone) it.
The analysis of accidents (and/or adjectives) presents an atypical interpre-
tation of traditional definitions:
The accident is what adheres (adest) and does not adhere (abest), i.e. ac-
cident is the expression that adheres to the subject, namely, to the sub-
stantive expression, i.e. it joins with i.e. fits in (the subject) co-signifying
it, and does not adhere to it, that is, it distances itself from that substan-
tive expression in the way of signifying, as it does not signify substantially
like that, i.e. it [the adjective] is far from the substantive expression to
which it adheres.14
Thus, the accident is not – as tradition dictates – the property that an indi-
vidual may or may not have, but something that has certain features in com-
mon with common nouns. The analysis of the adjective is not thorough, and
we can fill the gaps according to the instructions provided by Priscian (and
his commentators), who uses the concept of the ‘mode of signifying’ to dif-
ferentiate expressions that share the same meaning category. It is not plausible
that Garland is simply ignoring the traditional definition of accident, rather, it
seems more likely that the perspective of lectio nominis leads him to reformu-
late the canonical definition.
14
Dialectica, 10, 25–11, 1: ‘Accidens est quod adest et abest, idest illa vox appellatur accidens
que subiecto, idest substantive voci, adest, idest convenit consignificando ei, et abest,
idest longe est ab illa substantiva voce in modo significandi, quia non significat substan-
tialiter, sicut illa, vel abest, idest longe est a substantiva voce cui accidit.’
The ‘ Nominal ’ Stance 123
[Equivocal expressions] are not subjects, because they are neither first
nor second substances, […] every first and second substance is an ad-
equate subject for some accident. I say it is an adequate subject when
some expression substantially supports some accident equal to it, in such
a way that, being destroyed, it destroys the same accident, as the first
substance, which is this man, being destroyed, destroys this whiteness
together with itself.15
This text presents the same ambiguity that we have already encountered: on
the one hand, adjectives are expressions that are based on substantives; on
the other hand, the example of first substance is not formulated in metalin-
guistic terms. The example concerns the lower portion of the tree of predica-
tion; one might venture to pose the hypothesis that the tree whose nodes are
linguistic expressions corresponds, in some sense, to a set of specific objects,
for example:
15
Dialectica, 12, 6–13: ‘(Equivoca) subiecta non sunt, quia neque prima neque secunda
substantia sunt, […] omnis substantia, sive prima sive secunda, sufficiens subiectum est
alicui accidenti. Sufficiens autem subiectum esse dico quando aliqua substantialis vox
sustinet aliquod accidens par sibi quod destructa destruat, sicuti hec prima substantia
que est hic homo, destructa destruit hanc albedinem parem sibi.’
124 Chapter 5
‘animal’
↓
‘man’
↓
‘Plato’ ‘Socrates’ ‘Cicero’
⇓ ⇓ ⇓
Plato Socrates Cicero
This augmented tree – or the system consisting of the tree of predication and
the interpretation of terms16 – gives metaphysical color to the nominal inter-
pretation; it could correspond to an ontology of individual entities, linguistic
and extra linguistic. I say ‘could’ because Garland does not believe he needs to
explain the shift from use to mention of terms. In the text above he also speaks
of a certain whiteness which is destroyed along with the body that supports it.
This suggests that categories different from substance can be depicted in the
same way. This is confirmed by a remark on ‘to do’ and ‘where’:
It is asked why Aristotle considers the infinitives ‘to do’ and ‘to be affected’
(facere et pati) as most general rather than other verbs […] it seems that
he did so because there was no other way of giving individuals […] as in
‘to do this is to do’.17
‘Where’ and ‘when’ cannot be called highest genera except by adding
the verb ‘to be’, not being able to be predicated by themselves of their spe-
cies and individuals […] in the sentence ‘to be in Rome is to be in a place’
it is clear that ‘to be’ is added to Rome. 18
16 Neither possibility is useful in defining the relation outlined in the figure above between
a syntactic and a semantic tree of predication: the former only allows for a mixed tree,
or perhaps for replacing singular terms with singular entities, the latter only allows for
defining a semantic relationship between the syntactic tree nodes and a set of individual
entities.
17 Dialectica, 37, 27–33: ‘Queritur cur potius hec infinitiva, scilicet ‘facere’ et ‘pati’ et ‘habere’,
voluit esse generalissima Aristoteles quam alia verba […] Quod satis iusta de causa fecisse
videtur, hac scilicet quoniam alio modo individua his dare non posset […] ut dicere pos-
semus ‘hoc facere est facere’.’
18 Dialectica, 38, 33–39, 5: ‘Non enim aliter ubi et quando generalissima dici possent nisi eis
‘esse’ adiungeretur, cum de suis speciebus et individuis per se predicari non possent […]
In hac propositione: ‘esse Rome est esse ubi’ plane ostenditur [esse] Rome esse adiunc-
tum.’ Thus, being in Rome is an individual of the ‘where’ category. Garland complicates
things (if further complication were possible) to form a general proposition ‘every being
The ‘ Nominal ’ Stance 125
Substances are divided into first and second. The former are called substan-
tial individuals, the latter are called species and substantial genera. Speaking
of the individual this man, Garland says this refers to the man now present,
suggesting that it is the proper noun of that individual. Likewise the individ-
ual Socrates refers or leads by imposition to a person that cannot be currently
present. Conversely, shortly after individuals become non-linguistic objects, as
they admit of accidents and ‘pass them to second substances’, indeed – the
author says – if Socrates is white, (some) man is white too.
Quantities are what we expect them to be: lines, surfaces, bodies, places and
moments in time. The same things, however, can be interpreted as linguistic
expressions. A small bridge is thrown across the domain of expressions and
things: the body – according to Garland – belongs to the category of substance
in that it means the pure being of a thing; time – considered as vox – is predi-
cated of the present, past, and future. The passage would be more readable
if Garland provided us with some semantic framework. Among quantities,
speech is particularly interesting: it is defined as a set of phonemes (collectio
aeris). The language-metalanguage ambiguity is maintained, as speech (vox
complexa) is a linguistic element but also a physical object, considered as a
set of phonic elements uttered in succession. In general there is a second am-
biguity between an interpretation of a vox as linguistic type to which a certain
semantic function is assigned, and as phonic utterance having certain physical
features.
Relations are characterized as expressions that imply something else, in the
sense that the knowledge of a relative expression implies that of the other, as
when one says ‘he who knows what the master is also knows that with respect
to which it is said, i.e. what the servant is.’ The example could be paraphrased
in such a way to clear all terms not mentioned and get to something like this:
‘he who knows what is referred to by ‘master’ also knows what is referred to by
the relative term, i.e. ‘servant’’, but this formulation is unnatural and forced.
In some cases the non-mentioned terms cannot be eliminated; talking about
the simultaneity of relatives, Garland excludes the possibility that knowledge
and science are such, because ‘that thing of which one can have science comes
before the science that one has of it.’ The discussion of relations is complicated
by the fact that Aristotle does not explain very well what they are, nor what
properties relative things should have. In particular, the connection between
relations and related objects is not completely clear.19 From this point of view
some of Garland’s incongruous remarks about the elimination of related
things may be justified.
Quality is an expression said of certain quality-having things (quales). There
exist four kinds of expressions of quality which are subordinate to the noun
‘quality’, namely, the set of those expressions that mean a habit or a disposi-
tion, an ability or inability, a passion, or a form or figure. Garland explains the
significance of the nouns of the first kind in these terms: ‘those expressions that
mean habits and dispositions, i.e. the expressions ‘habit’ and ‘disposition’’.20
What Garland says on meanings is actually to be taken with extreme caution,
because – as in this case – the meaning of an expression is often composed of
other expressions. The reading of the chapter on quality increases the sense of
discomfort one feels each time Garland shifts from the plane of expressions
to that of (signified) things; this section of the Dialectica presents the greatest
difficulties, as one can see from the following table:
Perspective
19 Garland however cleverly points out a difficulty that comes from some Aristotelian ex-
amples (such as wing vs. winged): cf. Dialectica, 31, 1–7.
20 Dialectica, 33, 10–11.
The ‘ Nominal ’ Stance 127
Perspective
cont.
Perspective
a Dialectica, 33, 17–34, 5. In the column ‘semantic’ I only enter sentences that explicitly refer
to the relationship between expressions and meanings.
b Dialectica, 34, 6–12.
c Dialectica, 34, 13–35, 17.
d Dialectica, 35, 18–36.
When reading these texts, what first comes to mind is that the semantic ter-
minology is somewhat archaic, Garland seems to want to connect a semantic
interpretation to a purely nominal one, but he does not have the relevant ex-
pertise, nor is he entirely convinced of having to give a semantic framework.
The middle column is the showcase of Garland’s experiments, and sometimes
The ‘ Nominal ’ Stance 129
there are statements which can be traced throughout the past and future his-
tory of medieval semantics (from Boethius to Abelard), and at other times one
encounters provocations such as the word ‘honey’ that sweetens the mouth.
Garland may have lacked relevant information on Boethius or Priscian; how-
ever, the fact is that the theories of this period are very elementary, and simply
establish a link between terms and meanings referred to by derivative lexical
forms (such as ‘being man’, or ‘whiteness’); the first column of the table con-
tains formulations of this kind.
One wonders whether statements such as ‘‘hot’ implies a passion accord-
ing to its meaning’ are more intuitive than something like ‘‘heat’ means heat’.
It seems more likely that Garland considered the theses of a semantic theory
in the style of Boethius or Priscian to be either philosophically problematic or
barely informative. The column on the right shows terms used in seemingly or-
dinary contexts, perhaps pre-theoretical in the sense that when we say some-
thing like ‘someone is blushing with shame’, we do not commit ourselves to the
claim that shame – understood as an abstract entity – exists. Here too, how-
ever, things are somewhat uncertain: unless metalinguistically paraphrased,
‘this rationality is constitutive of Socrates’ sounds like a metaphysical proposi-
tion which entails acknowledging the existence of some particular property
(Socrates’ rationality).
The passages on meaning in the Dialectica reveal different degrees of ad-
herence to the Boethian standard approach. In the case we have discussed,
Garland’s terminology deviates from the standard, but perhaps retains a realist
point of view on meaning. In some cases meanings are seen as real entities;
in others as linguistic expressions which have a connection with the terms of
which they are meanings. It seems appropriate to explore these different as-
pects by considering some texts that contain a higher concentration of seman-
tic terminology.
Adjectives are an important issue discussed in the eleventh and twelfth cen-
tury. According to the standard interpretation, adjectives refer to things as well
as to properties. In keeping with this tradition, Garland argues that ‘white’ can
be understood for (accipi pro) its form, i.e. whiteness, or for that which has
whiteness, that is, white things. The analysis of adjectives is part of a wider
discussion of ambiguity, which, at least for a portion of the road runs parallel
to the traditional one:
21 This passage is probably a transcription of an oral lesson in which Garland elucidates
some written text.
22 Dialectica, 14, 15–23: ‘(equivocum est ens) ideo scilicet quia vadit ad diversa significata
per diversas diffinitiones et significationes, sicuti ‘canis’ equivocum. Ut hic patet: ‘hic
canis est canis in designatione latrabilis animalis’; hic cum una diffinitione venit ‘canis’
ad significatus suum, quod est latrabile animal; hic vero per aliam: ‘hic canis est canis in
designatione celestis sideris’; iterum per aliam: ‘hic canis in designatione marine belue’; et
sic propositio potest esse vera vel falsa diversis respectibus secundum predicatum quod
equivoce vadit ad diversa significata.’
The ‘ Nominal ’ Stance 131
It seems to us that there is nothing wrong if […] the species means it-
self and its genus, and some individual means himself, his species, genus
and also the highest genus; […] it could be shown in this way: if air is a
kind of body and ‘air’ means the expression ‘body’ and ‘substance’, then /
body/, which is the genus of /air/, means itself and a substance, because
all that is meant by the species as species is signified by its genus. The
individual who is ‘this air’ means itself. This is proved along the following
lines: all that is meant by the species is meant by some individual under
the species; if /air/ is the species of the individual which is this air, /air/
means the utterance ‘this air’, therefore the utterance ‘this air’ is signified
by some individual, and then by this air.24
This position is clearly nominal: genera, species and individuals are considered
linguistic expressions. One may ask what is the degree of the aggregation of
meaning: for example, if there is a principle according to which any predicate
may appear as a node of the predication tree, what should the status of mean-
ing of ‘white’ be – if there is one – when it is predicated with and without a
common noun? Garland has no advice to give, he only states that the ‘species
/air/ means the genus /body/ because /air/ is a species of /body/,’ somehow
binding the meaning to the relationship of predication.
We can extract from the passage above the following thesis regarding the
meaning of general expressions:
24
Dialectica, 20, 28–21, 5: ‘Nobis autem nullum inconveniens esse videtur si […] speciem
seipsam significare et genus suum et quoddam individuum seipsum significare et speciem
suam et genus suum, etiam ipsum generalissimum […] Quod sic probatus: aer est species
corporis et aer significat hanc vocem que est ‘corpus’ et hanc vocem que est ‘substantia’;
ergo corpus, genus aeris, significant seipsum et substantiam, generalissimum, quia quic-
quid significat species inquantum et eo respect quo est species, significat genus suum.
Individuum vero quod est hic aer significat seipsum; quod sic probatur: quod significatur
a specie, significatur ab aliquot individuo illius specie; si aer est species huius individui
quod est hic aer, et aer significat hanc prolationem que est hic aer; ergo hec prolatio que
est hic aer significatur ab aliquo individuo aeris, ergo ab hoc aere.’
The ‘ Nominal ’ Stance 133
a. ‘body’, which is the species of ‘air’, means itself and the substance
b. ‘This air’ means ‘this air’
A term that labels a node of the tree of predication, in some sense, refers to
all of the nodes that one encounters going back up the tree to the apical end.
However, this interpretation contrasts with the first claim above. An exten-
sional interpretation of meaning should hold symmetrically:25 that is, all the
elements belonging to the species (ultimately considered its meanings), be-
long to the genus.
Let us denote the set of meanings of ‘man’ in the following way:
25 I partially agree with Tweedale, who speaks of an extensional perspective when com-
menting on the text above. I do not, however, agree with the parallel he suggests between
the relation of meaning in Garland and denotation in J.S. Mill. Unless one always finds
a way of translating Garland’s statements on nominal meaning into statements about
things denoted – which is no easy task, as we have tried to show – the comparison is
doomed to fail. Cf. M. Tweedale, ‘From the Late Eleventh Century to the Time of Abelard,’
Cap.7 di P. Dronke (ed.), A History of Twelfth-Century Western Philosophy, Cambridge UP,
Cambridge 1992, 200.
134 Chapter 5
where the expressions to the right of the square bracket represent what is over
the species, intended as a node of the predication tree, while the expressions
to the left represent the ‘meanings’ under or coinciding with the species (as
Garland says, what is meant by the species as species). How should we view
these meanings? According to Garland’s assertions in the text, there seems to
be no ontological difference between meaningful expressions and meanings:
both are linguistic expressions.
Garland also claims that individuals mean themselves; a claim to be kept in
mind while looking down the tree of predication or at the left bracket in the
formula which follows:
man
Socrates
I do not believe, however, that Garland can reject the principles of predication
set out in 5.1. For example, in the case of /man/ and the genus /animal/ we have
the following instance of the second ‘radical’ nominal principle:
26 See in this regard M. dal Pra’s ‘introduction’ to P. Abelardo, Scritti di Logica, ed. M. dal Pra,
Milano 1969; C. Mews, ‘On dating the works of Peter Abelard’, in Archives d’histoire docri-
nale et litterarire du Moyen Age, 1985 (LII), 73–134; J. Marenbon, The Philosophy of Peter
Abelard, Cambridge University Press, Cambridge 1997, pp. 37–40
27 On the literary genre of the commentaries see, inter alia: J. Marenbon, ‘Medieval Latin
Commentaries and Glosses on Aristotelian Logical Texts Before c. 1150 AD’, in J. Marenbon,
Aristotelian logic, Platonism and the Context of Early Medieval Philosophy in the West,
Aldershot, Ashgate 2000, item II; S. Ebbesen, ‘Medieval Latin Glosses and Commentaries
on Aristotelian Logical Texts of the Twelfth and Thirteenth Centuries’, in C. Burnett (ed.),
Glosses and Commentaries on Aristotelian Logical Texts. The Syriac, Arabic and Medieval
Latin Traditions, The Warburg Institute, London 1993, 77–97, 129–147; S. Ebbesen, ‘The
Aristotelian commentator’ in: John Marenbon, ed., The Cambridge Companion to Boethius,
Cambridge University Press, pp. 33–55, Cambridge 2009.
The ‘ Nominal ’ Stance 137
the texts of Porphyry and Aristotle are to be read somehow like treateses on
logical syntax, is documented, among others, by the following passages:
Porphyry’s intention in this work is to deal What the first substance means, that is,
with the six expressions. the expression ‘Socrates’.
The expression to which the species is The substance is the expression that
submitted is called genus. admits contrary qualities.
The expression submitted to the genus is The substance, that is, the expression
called species. meant by the expression ‘substance’.
That noun is called most general (or most Expressions that are called determinate
special). quantities.
That expression is called difference. Quantity is the expression placed in the
Accident, i.e. that expression called category of quantity.c
accident.b
a To be honest the first part of what remains of the glosses on Categories does not contain clear
signs of the author’s belonging to the nominal sect. In fact, the first part speaks of species,
genera and substances (first and second) without adding the clause ‘that is, the expression …’
Certainly, these are always things that are predicated and predication is likely to be under-
stood as a relation between linguistic objects. However, it is a fact that for a significant part
of the commentary the author does not commit himself to holding an exegetic position; in
the second part we find clearer elements of the lectio nominis that are interwoven with an
(attempted) semantic analysis.
b P. Abelardo, Scritti di Logica, Editio super Porphyrium (=Ed. Sup. Por.); 3, 1–2: ‘Intentio
Porphyrii est in hoc opere tractare de sex uocibus’; 8, 31: ‘Illa uox dicitur genus cui supponitur
species’; 12, 28: ‘illa uox dicitur species quae supponitur generi’; 14, 15–16: ‘Illud nomen dicitur
generalissimum (specialissimum)’; 29, 7: ‘illa uox dicitur differentia’. 36, 13: ‘Accidens, scilicet
illa uox dicitur accidens’.
c P. Abelardo, Scritti di Logica, Glossae in Categorias (=Gloss. In Cat.); 54, 23–24: ‘illud quod sig-
nificatur a prima substantia, scilicet ista uox quae est Socrates’; 57, 32–33: ‘substantia, id est
vox significata per hac voce quae est substantia’; 58, 1–2: ‘substantia est uox susceptibilis con-
trariorum’; 56, 41–57, 1: ‘illarum uocum quae determinatae dicuntur quantitates’; 62, 25–26:
‘quantitas est uox proprie ordinata in praedicamento quantitatis’.
This way of seeing things does not come as a great surprise: in fact the same
commented authors sometimes express themselves in these terms. Especially
when the intention is to write an introductory treatise to logic, taking a met-
alinguistic point of view can be a good strategy. However, this is only the
138 Chapter 5
28
Ed. Sup. Por., 20, 9–11: ‘Participatione enim speciei, id est hominis, plures homines, id est
Socrates et Plato, sunt unus in hac uoce quae est homo’.
29
Ed. Sup. Por., 31, 24–32: ‘Si aliqua uox praedicatur de alia uoce in quid (quale), non sequi-
tur quod constituat eam materialiter (formaliter)’.
30
Gloss. In Cat., 55, 1–5: ‘Homo ponit animal informatum qualitatiua uoce, id est rationali-
tate in sui constitutione et animal ponit substantiam suam materiam informatam quali-
tatiua uoce, id est animatione et sensibilitate in sui constitutione’.
The ‘ Nominal ’ Stance 139
It is remarkable that the lectio rei coincides to the letter with the semantic side
of the lectio nominis: discreet = meaning of ‘discreet’. The same things return in
the Logica Ingredientibus, where the statement of principle that the real plan
must be distinguished from the nominal plane, is followed by the introduction
of a third semantic exegesis, according to which the ‘things’ the logical authori-
ties talk about have to be understood as meanings. Much of Abelard’s work in
the later commentaries is to try and justify the introduction of entities signi-
fied by general terms. In the Literal Glosses one should not expect to find this;
however semantics – as we will see – sometimes appears behind the curtain of
the lectio nominis.
As to properties, in one case Abelard expresses himself enigmatically, since
on the one hand he seems to speak of common properties, on the other of the
simple sharing of a noun:
The property of the individual will not be in other individuals; but the
properties of the individual man, I mean the common and special man,
will be the same in many things; in reality, in all, and this appears in the
fact that, or rather because of the fact that, they are men, that is, because
man, this expression, is suitable for all men.32
Here we do not find the exaggeration we would expect, i.e. that the common
man, understood as noun, is in several things, but the rather neutral argument
that man’s common properties are in several things as several things share
the noun ‘man’. Other texts have the same ambiguity, as if Abelard wanted to
avoid Garland’s extravagances, without distancing himself from the narrow
nominal reading. Something similar occurs in passages where semantic terms
31
Gloss. In Cat., 63, 31–40: ‘Discretum duobus accipitur modis: scilicet in designatione rerum
et uocum; et acceptum in designatione rerum sic definitur: discretum est mensura adia-
cens substantiis secundum quam ipsae substantiae dicuntur discretae succedenter aut
permanenter […] Et acceptum in designatione uocum talem habet definitionem: discre-
tum est uox significans mensuram adiacentem substantiis secundum quam ipsae sub-
stantiae dicuntur discretae succedenter aut permanenter.
32
Ed. Sup. Por., 22, 26–30: ‘Proprieta<te>s indiuidui non erunt in aliis; sed proprietas indi-
uidui hominis, hominis dico communis, id est specialis, erit eadem in pluribus; sed magis
possum dicere, in omnibus, et hoc apparet in eo, id est ex eo, quod homines sunt, id est
quod homo, ista uox, conuenit omnibus hominibus’.
140 Chapter 5
are introduced. Let us see what things are considered as meanings of logical
expressions and what kind of logical terminology is employed (the first and
second text are from the commentary on Porphyry, the third from the com-
mentary on the Categories):
This work concerns these six expressions and their meanings […] here we
teach how to find arguments to prove any question that is asked about
these six expressions and their meanings.33
Porphyry shows that the expression ‘man’ is given with different mean-
ings (e.g. Socrates and Plato).34
When ‘quantity’ means the things themselves, that is, the same
measurements and expressions placed in the category of quantity, it is
equivocally said of those things, since any expression that signifies things
signified by a different highest term is equivocal; but quantity signifies
the line, the surface, and other things of this type, which are signified
by substance that is a highest term, and means measures adjacent to
substances.35
In the last passage Abelard seems to support the rather surprising thesis
that every quantitative term when understood in a real sense means things
placed under the category of substance. I think he really just wants to say that
‘quantity’ (as meta-term) stands both for quantitative things and for quanti-
ty expressions; the latter, as sound materials, are ‘things’ to be placed in the
category of substance. Of course, the last phrase ‘substance that is a highest
term … means measures adjacent to the substance’ requires some additional
interpretative virtuosity.
The terminology is poor: Abelard only uses the verb ‘significare’ (which
I translate ‘to mean’ or ‘to signify’ depending on the context) in the generic
sense and in the more precise sense of ‘refer to’. In the first decades of the
twelfth century, the semantic terminology used by the masters is quite varied;
33
Ed. Sup. Por., 3, 20–4, 2: ‘Huius operis sunt materia istae sex uoces et earum significata [….]
hic docemur invenire rationes sufficientes ad probandas quaslibet quaestiones factas de
istis sex uocibus et de significatis earum’.
34
Ed. Sup. Por., 20, 5–6: ‘Hanc uocem quae est homo ostendit dari diuersis significatis
Socrates et Plato’.
35
Gloss. In Cat., 62, 31–63, 4: ‘Cum quantitas significet ipsas res, id est ipsas mensuras et
uoces proprie positas in praedicamento quantitatis, est aequiuocum ad illa, quia quae-
cumque uox significat res significatas a diuerso generalissimo est aequiuoca; sed quanti-
tas significat lineam et superficiem et consimilia, quae sunt res significatae a substantia
quae est generalissimum et significat mensuras adiacentes substantiis’.
The ‘ Nominal ’ Stance 141
one may therefore wonder whether the author was familiar with this terminol-
ogy or wanted to consciously speak of meaning only in terms of reference. A
possible (and I think unique) counter example comes from the following text:
Genera and species determine the quality of the substance, as they mean
the substance, that is, a certain matter, i.e., something that is a unity; (they
mean) its own qualified matter, that is, conformed in its constitution.36
The quality that informs here is the object of determinative meaning. It could
be the usual way of presenting the determinatio as a relationship between ab-
stract expressions and meanings (‘man’ → being man, ‘white’ → whiteness);
however, ‘determinatio’ is used in some places with an extensional meaning
and the reference to the constitution leads us to link this text with those cited
on the constitution of nominal definition.
A block of texts in the commentary on the Categories contains a particular
use of ‘to mean’. We have already noticed this use in Garland. Here are two
examples:
36
Gloss. In Cat., 55, 31–34: ‘Genus et species determinant qualitatem circa substantiam quia
significant substantiam, id est materiam quamdam, id est quodque unum, suam propri-
am qualem, id est informatam qualitate in sui constitutione’.
37
Gloss. In Cat., 57, 12–17: ‘aliqua uox significata ab hac uoce quae est substantia non suscipit
maius et minus in suo proprio uocabulo, respectu suppositorum. Non enim potest dici si
est magis uel minus homo; sed homo in suo significante potest recipere magis et minus,
sic homo est magis substantia quam animal’.
38
Gloss. In Cat., 61, 18–22: ‘Oratio namque et opinio dicuntur esse susceptibilia contrari-
orum non in eo quod ipsa recipiant aliquid accidentaliter, sed in eo quod sit facta aliqua
passio, id est aliqua accidentalis permutantio circa alterum, id est circa rem significatam
a suo significato’.
142 Chapter 5
39
Gloss. In Cat., 58, 10–28: ‘Susceptibile contrariorum est quando aliqua uox suscipit in
diuersis temporibus duo contraria circa ipsam uocem […] albedo et nigredo quae sunt
causae albi et nigri non sunt diuersae ab illa re quae significatur ab hac uoce quae est hic
color, quia eadem res quae significatur ab albedine et nigredine significatur ab hac uoce
quae est hic color, ens indiuiduum coloris speciei. Ideo hic color ens indiuiduum coloris
speciei uidebatur suscipere album et nigrum quia hic color ens indiuiduum albedinis for-
mae aut nigredinis formae par est in significatione ad hoc coloratum et hoc coloratum
ad hoc corpus et hoc corpus suscipit album et nigrum; et ita hic color ens indiuiduum
albedinis formae aut nigredinis formae suscipit album et nigrum’.
The ‘ Nominal ’ Stance 143
Therefore, this color admits white and black. And since all this applies to
‘this sentence’ (i.e. this vocal expression) as it admits opposite determi-
nations at different times, this vocal expression admits the true and the
false at different times not according to its modification but according to
the modification of what is signified by its meaning, that is to say, accord-
ing to the modification of Socrates, which changes from sitting to stand-
ing. Alternatively, the thing, that is Socrates, is meant by the sentence
‘Socrates sits’, which is signified by the expression ‘sentence’.40
This statement admits the true/false: ‘this statement’ said in a certain cir-
cumstance→ ‘Socrates sits’ → Socrates (who is sitting)
This color admits whiteness-darkness: ‘this color’ said in a certain cir-
cumstance → this color white
40
Gloss. In Cat., 58, 28–43: ‘Et ita hic color suscipit album et nigrum. Et quia hoc totum
habet haec oratio, ista uox quia suscipit contraria in diuersis temporibus ista uox suscipit
uerum et falsum in diuersis temporibus non secundum mutationem sui, sed secundum
mutationem rei significatae a suo significato, id est secundum mutationem Socratis, mu-
tati de sessione in stationem, aut e conuerso, quae res, id est Socrates, significatur ab hac
oratione quae dicit: Socrates sedet, quae oratio, id est Socrates sedet, significatur ab ista
uoce quae est oratio’.
144 Chapter 5
Those things, that is, the individuals, who are in the ascending series be-
fore the lowest species, are said to be ‘ascending’ that thing, i.e. genera
or species, that is the meanings of those expressions. […] The nature of
the individual things is always to ascend, by accepting the predication of
the higher things, as well as, inversely, the nature of the general ones is to
descend by division.41
41
Ed. Sup. Por., 16, 33–17,2: ‘ea, scilicet indiuidua, quae sunt in ascensu ante specialissima,
ipsa dicuntur ascendentia hanc rem, scilicet genera et species, id est ipsa significata
harum uocum; ….Natura enim indiuiduorum semper est ascendere suscipiendo praedi-
cationem superiorem […] sicut e contrario natura generalissimorum est descendere per
diuisionem’.
The ‘ Nominal ’ Stance 145
Genera and species are predicated of things that lie beneath them and at the
same time mean real things. As for the principle of predication, what we find is
a more or less literal quote of the texts, as in the following instance:
If the tree is to be drawn according to the schema above, evidently we can not
move away from a version of predication principles of the kind:
The texts quoted above and the mention of the causes would not lead me to go
any further; the suggestion is that the modern commentator stops where the
young Abelard (?) stopped, with a doubt as to whether the reading comple-
mentary to the lectio nominis can lead to a moderate semantic realism or ex-
treme metaphysical nominalism. At the same time, one may believe that the
seeds of the third semantic lectio sown by Garland and the author of the Literal
Glosses have sprouted in later commentaries by Abelard.
42
Ed. Sup. Por., 21, 22–27: ‘Vere de quibus praedicatur species, facit superiora predicari,
quia si uerum est Socratem dicere hominem, <hominem> autem animal, sequitur istud
uerum: Socratem dicere animal; et si uerum est Socratem dicere animal, animal <autem>
substantiam, uerum est Socratem dicere substantiam’.
chapter 6
The theories that identify individual and universal things draw on Boethius’
second commentary on Porphyry. We have seen that Boethius believes similar-
ities are present in things perceived by the senses and ‘become’ ( fiunt) individ-
uals; from a different point of view, the similarities are considered as universal
species and genera. The critical point concerns the interpretation of the verb
‘fieri’: there exists a certain conflict between an ascription of mental attitudes
(attentiones) and the identification in re of general properties. In the theory
that we present in this chapter the psychological aspect is secondary; what
matters is the actual identification of species-genera with individuals. The
first part of the manuscript BN 17813 (pages 1–16, which we will call Questiones
from its incipit) takes the form of a summary of the salient passages of the
second Boethian commentary on Porphyry. It can be considered an archaic
work, since all the stylistic elements usually characteristic of commentaries in
this period are absent, such as critical expositions of the different positions on
the problem of universals, detailed discussions of various theses, and familiar-
ity with logical-semantic terminology. Curiously, the author apparently holds
the identity theory (that universal and particular coincide), which was prob-
ably formulated sometime between 1110 and 1130 as a response to the thesis of
William of Champeaux.
It may be that the anonymous author simply wanted to expound some
key passage of Boethius’ work, and did not notice the gradual shift from the
literal commentary to an autonomous survey, for which a deeper and more
well-documented analysis was evidently required. In this case, perhaps, the
Questiones serve as a link between the Boethian theory of similitudo and that
of identity. The fact that in the manuscript the commentary with which we are
dealing immediately precedes that on genera and species attributed to Walter
of Mortagne,2 however, indicates another possibility. Giving importance to the
1 For a description of the manuscript see B. Hauréau, Notices et extraits de quelques manu-
scripts latins de la Bibliothèque Nationale, II, Paris 1891, 298–333; J. Dijs, ‘Two Anonymous
12th-Century Tracts on Universals,’ Vivarium, 28, 1990, pp. 85–113.
2 This attribution is based on the testimony of John of Salisbury, who speaks of Walter as the
leader of a sect that claimed that species and genera are individuals sharing a certain sta-
tus. The author of the text of BN 17813, 16VA–19RA claims the same things, demonstrating
presentation of the thesis of identity which appears in both texts, one might
think that the author of the commentary on Boethius were Walter himself,
who considers it appropriate to preface the detailed explanation of his thesis
with a short, concise commentary on Boethius’ text. According to this interpre-
tation the initial sentence ‘Ingredientibus logicam liber Porphyrii, quia primis
legendus occurrit, inter huius libri initia ipsius logicae ortus praelibandus vide-
tur’ should not be taken as a procedural formality, but as a way to separate a
first recognition of the problem of universals from a second and more careful
critical examination.
Whatever the case, in order to check for deviations I think it is useful to
juxtapose some significant passages of the anonymous commentary to the
Boethian text. The solution of the Porphyrian questio is introduced in the fol-
lowing way:
This question is resolved by Boethius, Among the issues we have talked about,
following Alexander, in this way: genera this one now arises, which we solve
and species certainly have their being following Alexander […] There are
in individuals, but do not exist in more things which have their being in
themselves separated from individuals.a others from which either they cannot
be completely separated, or, if they were
separated, would for no reason subsist.b
a Haureau, op. cit., 295: ‘Haec quidam questio a B., Alexandro consentiente, hac ratione solvi-
tur: genera et species esse quidam habent in suis individuis, non tamen habent existere per
se in sua semplicitate, discreta ab individuis.’
b In Isag. II, 164, 3–16: ‘Haec quidem est ad praesens de propositis quaestio, quam nos
Alexandro consentientes hac ratiocinatione soluemus […] Sunt enim plura quae in aliis esse
suum habent ex quibus aut omnino separari non possunt aut, si separata fuerint, nulla ra-
tione subsistunt.’
considerable dialectical skills. The editor recognizes that such evidence is not entirely con-
clusive. The dating of the text is uncertain (Dijs proposes the 1110–1130 period), and so is the
date of birth of Walter of Mortagne (see W. Robl, Walter von Mortagne. Leben und Werk:
http://www.abaelard.de/abaelard/030033mortagne.htm).
It is merely for the sake exposition that I attribute to Walter the theory supported in the
treatise on the status of genus and species. On Walter’s life and attribution of De Statu see
C. Tarlazzi, Individui universali. Il realism di Gualtiero di Mortagne nel XII secolo, Brepols,
Turnhout 2017, par. 1.1–2, 1.7.
148 chapter 6
Boethius and the anonymous author claim that genera and species have their
being in individual things. According to both, incorporeal nature does not exist
separate from the bodies in which it is immersed.
The anonymous author follows Boethius quite closely in describing the pro-
cess of abstracting a geometrical entity such as a line from the surface in which
it is located; he introduces the verb ‘existere’ instead of the Boethian ‘subsistere’
to contrast the ontological status of the line compared to the body:
So far, the differences have perhaps been somewhat insignificant, but this
changes when the anonymous author introduces Boethius’ solution:5
Genera and species … cannot be So even for genera and species, i.e.
without individuals, as they are but as to singularity and universality,
individuals considered from another the subject is unique, but in a way
point of view. The same thing is in regarded as universal when it is
fact individual, species and genus, thought, in another as singular when
considered in different ways, as well it is perceived in the things in which
as Socrates is both man and animal. it has its being.5
They are therefore in some way
things in corporeal bodies, yet they
are grasped by the intellect beyond
what is perceived by the senses in
3 Haureau, op. cit., 296: ‘si quis igitur abstrahat lineam a corpore, quia essentiam rei existentis
concipit, verus est ejus intellectus.’
4 In Isag. II, 165, 17–166, 5: ‘eam (lineam) mente capimus quasi praeter corpora sit .. qui uero
id (intellectus) in diuisionibus et abstractionibus … efficit, non modo falsus non est, uerum
etiam solus id quod in proprietate uerum est inuenire potest.’
5 In Isag. II, 167, 3–7: ‘Ita quoque generibus et speciebus, id est singularitati et uniuersalitati,
unum quidem subiectum est; sed alio modo universale est cum cogitatur, alio singulare cum
sentitur in rebus his in quibus esse suum habet.’
Walter of Mortagne and the Identity Theory 149
Here the7anonymous author presents the identity thesis: genera and species
‘are nothing more than individuals’; ‘the same thing, otherwise considered, is
individual, genus and species.’ In Boethius’ text the assertion found to be
6 Haureau, op. cit., ibid.: ‘genera et species … non possunt esse sine individuis, quia nihil sunt
aliud quam individua, sed alio modo considerata. Eadem enim res et individuum et specie
et genus est diversis modis considerata; sicut eadem res et Socrates est et homo et animal.
Quando intelligitur formata animatione et sensibilitate, genus est; quando vero intelligitur
formata rationalitate et mortalitate, species est; cum autem intelligitur discreta et deter-
minata accidentibus, individuum est. Itaque eadem res est et singularis et universalis, alio
modo et alio modo considerata, quemadmodum eadem linea et cava et concava. His igitur
determinatis, omnis questio dissoluta est.’
7 In Isag. II, 166, 6–8: ‘Sunt igitur huiusmodi res in corporalibus atque in sensibilibus, intelle-
guntur autem praeter sensibilia ut eorum natura perspici et proprietas ualeat comprehendi.’
8 In Isag. II, 166, 8–14: ‘Quocirca cum genera et species cogitantur, tunc ex singulis in quibus
sunt eorum similitudo colligitur, ut ex singulis hominibus inter se dissimilibus humanita-
tis similitudo, quae similitudo cogitata animo ueraciterque perspecta fit species; quarum
specierum rursus diuersarum similitudo considerata, quae nisi in ipsis speciebus aut in
earum indiuiduis esse non potest, efficit genus.’
9 In Isag. II, 167, 7–8: ‘His igitur terminatis omnis, ut arbitror, quaestio dissoluta est.’
150 chapter 6
closest to this thesis is passage 167, 3–7, in conjunction with 166, 6–8. In the
former it is said that the same thing is regarded as universal when it is thought
and as singular when it is perceived; in the latter Boethius states that genera
and species are things in sensible things (res in corporalibus atque in sensibili-
bus). This may have been the starting point for the subsequent elaboration of
the identity thesis. As often happens with Boethius, however, things are not so
clear as the anonymous commentator would like us to believe.
One must initially wonder what the exact meaning is of the statement ‘one
thing only is subject to genera and species, i.e. to singularity and universality.’
If Boethius means that the same thing is genus, species and individual, why in-
stead he says that such a thing is subject to genera? On the other hand, it is true
that interpreting the subject thing as an individual, the next sentence of 167,
3–7: ‘(such a subject thing) is singular and universal’ brings us very close to the
identity thesis. The final sentence of Boethius’ text, however, prevents us from
considering ‘the unique thing’ as an individual, as this unknown entity must
have its being in things perceived by senses. For a correct interpretation, one
must consider the Boethian position as whole: as we saw in the second chapter,
similarities are the main players in the story told by Boethius. Once 167, 3–7 is
viewed in its context, the passage appears to be a reformulation of the main
thesis of the second commentary on Porphyry, according to which similarities
have different properties depending on the way in which they are observed.
As we have already said, what Boethius means is not entirely clear; it is thus
understandable why In Isag. II, 167, 3–7 gave rise to different interpretations, in
particular those of the anonymous author and Walter of Mortagne.10
10 What is stated here applies if we give importance to the relevant passages of the com-
ments on Porphyry as the main source for the theories we are examining. If, however, we
do not limit ourselves to these texts, there can be found in Boethius suggestions which
lead to the identity thesis, see for example In Isag. II, 191, 21–192, 13: ‘Tribus enim modis
unumquodque uel differre ab aliquo dicitur uel alicui idem esse, genere, specie, numero.
Quaecumque igitur genere eadem sunt, non necesse est eadem esse specie, ut si eadem
sint genere, differant specie. Si uero eadem sint specie, genere quoque eadem esse necesse
est, ut cum homo atque equus idem sint genere – uterque enim animal nuncupatur –
differunt specie, quoniam alia est hominis species, alia equi. Socrates uero atque Plato
cum idem sint specie, idem quoque sunt genere; utrique enim et sub hominis et sub ani-
malis praedicatione ponuntur. Si quid uero uel genere uel specie idem sit, non necesse
est idem esse numero, quod si idem sit numero, idem et specie et genere esse necesse
est; ut Socrates et Plato, cum et genere animalis et specie hominis idem sint, numero
tamen reperiuntur esse disiuncti.’; De Trinitate I: ‘idem dicitur vero tribus modis: aut ge-
nere ut idem homo quod equus, quia his idem genus ut animal; vel specie ut idem Cato
quod Cicero, quia eadem species ut homo; vel numero ut Tullius et Cicero, quia unus est
Walter of Mortagne and the Identity Theory 151
numero. Quare diversum etiam vel genere vel specie vel numero dicitur’ (De Trinitate, ed.
cit., 6, 18–24).
11 ‘If this hypothesis is correct, we could restrict the dates of the tract to a period between
the late 1110s and early 1130s’ (op. cit., 88). On the attribution of De Gen. Stat. to Walter, see
the previous paragraph.
12 See the previous note on the attribution and dating of the Treatise. It should be noted
that the dating of the Logica Ingredientibus itself is the subject of some debate: cf.
J.E. Brower – K. Guilfoy, ‘Introduction’, in J.E. Brower – K. Guilfoy (eds), The Cambridge
Companion to Abelard, Cambridge UP 2004, Cambridge, p. 7.
13 We call the second thesis TR3, to comply with the numbering of the thesis of realism
given in the chapter on realism.
14 We list below the criticisms of realism, the theory of non-difference and vocalism under
the abbreviations, ci.
152 chapter 6
c1. The donkey, as a universal, would have ubiquity in common with God, since
‘such a universal thing is in different places at the same time, for example,
here and in Rome.’ The formula ‘here and in Rome’ is interesting, as it is one of
the first records I know of a logical-grammatical difficulty which is present in
the Logica modernorum, particularly in some of the comments on Sophistical
Refutations15 and frequently in thirteenth and fourteenth century treatises on
suppositio.
c2. One wonders if the process that constitutes Socrates starting from man
is reversible. Walter considers it impossible that Socrates may coincide with
Plato once the differences that constitute them as individuals are taken away.
c3. One wonders if what happens to every man may happen to the species,
which is, in some sense, in that particular man. It is to be borne in mind that
for the thesis at issue the universal man is physically present in the particu-
lar individuals. Walter’s criticism is analogous to that of Abelard, but clearer.
I quote it entirely as – inter alia – it can help clarify the parallel passage of
Ingredientibus:
15 Cf. Fallacie Parvipontane, in De Rijk, Logica Modernorum, Vol I, Van Gorcum, Assen 1962,
599, 18–19; Summa Metenses, in De Rijk, Logica Modernorum, Vol II–1, Van Gorcum, Assen
1967, 455. A similar case is considered in the Summa Sophisticorum Elencorum: ‘quod est
Bononie, non est Rome,’ (Logica Modernorum, Vol I, 327, 19).
16 De Gen. Stat., Ed. Dijs (in op. cit.), 97, 12; ed. Romano (in Una soluzione originale della
questione degli universali nel XII secolo, Aracne, Roma 2007), 96, 14: ‘Item opponitur: cum
animal rationale mortale quod est in Socrate agate et patiatur, quaeritur utrum agant et
patiantur ipsa accidentia vel ipsa materia Socratis, vel utrumque. Sed de accidentibus
non concedunt quod patiantur, neque de utroque, scilicet de materia et de eius accidenti-
bus, quia tunc concederent de accidentibus, quod iam negatum est; concedunt ergo ma-
teriam pati, cum hoc vel illud esse debeat. Sed quotienscumque homo qui est in Socrate
Walter of Mortagne and the Identity Theory 153
agit vel patitur, et homo qui est in Platone agit vel patitur, cum sit eadem essentia, et sic,
<Platone> agente aliquid, agit Socrates et quaelibet alia substantia, et flagellato Socrate,
flagellabitur quaelibet alia substantia; quod est inconveniens, et etia haeresis.’
154 chapter 6
c7. Walter lends a counterargument to his opponents: ‘they perhaps do not say
that the mere animal, as such, is the sufficient foundation of accidents, nor
the animal formed from differences, rather the animal related to its inferiors.’17
Such a solution is reminiscent of the (Neoplatonic) distinction between the
universal in things and the abstract universal, and is developed in a manner
compatible with Boethius’ commentaries on Porphyry. In fact, this point marks
the transition from the criticism of TR1 to that of TR2 or TR3, since the latter
theses focus on the difference between simple animal and particular animals
(see the chapter on realism).
c8. When we say that the animal is a genus, the property of being a genus is
attributed to both the animal and the individual animal. Indeed, a realist who
supports TR2 and TR3 must at least accept that the names of species/genera
refer to entities of some sort. But if universal terms are authentic entity names,
the question arises whether what can be said of such entities can also be said
of subordinate individuals. The prescription of Categories 3 ‘what is said of
the species is said of individuals of the species’ contrasts with the existence of
predicates, for instance ‘being a species’, which are predicated exclusively of
universals. Walter notes this second aspect: if it were true that ‘genus’ is predi-
cated of some individual, it would be true that some animal is a genus, but the
realists themselves recognize that no animal is a genus. The alternative that
‘being a genus’ is only said of the simple animal is not feasible, since ‘the gen-
erality would be attributed to the simple animal qua simple, and in a way that
is not consistent with its being, since, according to this position, the animal
considered qua simple is nothing.’18 It is not clear, however, in this context in
17
De Gen. Stat., Dijs, 101, 19; Romano, 99, 24: ‘Sed forsitan neque dicunt animal simplex, ut
simplex, esse sufficiens fundamentum illorum accidentium, nec animal formatum, sed
animal circa sua inferiora.’
18
De Gen. Stat., Dijs, 101, 20; Romano, 100, 25: ‘oportet quod generalitas animali simplici at-
tribuatur, et ut semplici, et sic attribuatur animali eo modo quod non est, quippe animal
secundum hanc sententiam acceptum ut simplex nihil est.’
Walter of Mortagne and the Identity Theory 155
Walter observes that the different ways men have of looking at things have no
real impact, because they cannot make that which is, not be, or that which is
not, be. It should be noted that Walter is in a better position than Boethius,
since the former explicitly states that there are only individuals which may
be seen from different points of view, while the latter speaks of two things –
individuals and similarities or forms – whose difference can hardly be a matter
of perspective. The point is, then, that Socrates as Socrates is different from
every other individual, and as rational animal Socrates is non-diversified from
Plato. It is not easy to understand what this means. By saying that Socrates is
different from Plato, we normally mean that they are not the same individual,
or, in other words, that it is not true that Socrates is the same individual as
Plato. Given this, what does ‘Socrates is non-diversified from Plato’ mean? If
non-diversified means equal, clearly we have a contradiction with the banal
truth of ‘Socrates is not Plato.’
Walter does not succeed in making us believe that this is not a contradic-
tion, and it is only by eliminating the contradiction that his observations could
make sense. The first task is to differentiate identity from non-difference.
enough to justify the belief in universal entities. One can understand and share what this
author says, for example, Abelard is a philosopher who thinks the same way; my uncertain-
ty is due to the fact that universal expressions are usually considered to mean something
(property form, essence..) common to several particulars, whereas Walter’s universals are
by definition exactly the same as particular things. On the other hand, I completely agree
that the use of labels such as ‘moderate realism’ or ‘unrealism’ is completely arbitrary.
21
De Gen. Stat., Dijs, 103–104, 27, 29,30; Romano, 102–103, 31, 33, 35: ‘Quod qualiter sit per
diversas attentiones discernitur. […] Si quis ergo Socratem attendat tamquam Socratem,
id est in omni proprietate Socratis, inveniet eum cun nullo convenientem, potius ab om-
nibus differentem […] Sed si simpliciter attendatur Socrates non ut Socrates, id est non
in omni proprietate Socratis, sed in quadam, scilicet in eo quod est animal rationale mor-
tale, iam secundum hunc statum est differens et indifferens […] Ecce Socrates , secundum
statum hominis est species specialissima.’
158 chapter 6
Walter often uses the contrasting phrases ‘idem essentialiter’ and ‘indifferens’
to signify identity and non-difference respectively. In what follows we will
systematically translate the pair of expressions with ‘identical’ and ‘non-
diversified’, while we shall use the term ‘different’ (= ‘differens’) to mean ‘not
identical’. In doing so we want to suggest an interpretation of the text that al-
lows us to resolve the paradox of the identity of different things.
When considered according to the status of Socrates, Socrates is a different
individual from any other; on the other hand, when considered according to
the status of rational animal, he is a species and, as such, is non-diversified
from Plato, considered according to the same status.22 An individual thing can
be called species/genus insofar it has a certain property signified by a gram-
matical predicate:
22 The term ‘status’ is used by Abelard to indicate the meaning of a predicate, and it seems
to me that in this context the term can perform the same function. In the critical appara-
tus to his edition of Walter’s text, F. Romano notes that the term ‘status’ occurs 95 times
(p. 66). Dijs (op. cit., 92) rightly points out that it is incorrect to speak of a theory of status
as necessarily associated with so-called ‘nominalism’; in fact, different authors use the
notion of status in different contexts and with different purposes.
23 De Gen. Stat., Dijs, 104, 29; Romano, 103, 33: ‘Si simpliciter attendatur Socrates non ut
Socrates, id est non in omni proprietate Socratis, sed in quadam, scilicet in eo quod est
animal rationale mortale, iam secundum hunc statum est differens et indifferens; differ-
ens a qualibet alia re existente […] item indifferens est, id est consimilis cum quibusdam,
scilicet cum Platone et cum aliis individuis hominis, in eo quod in unoquoque eorum est
animal rationale mortale.’ I have translated the last sentence assuming that the editor’s
reading of the manuscript is correct; it would be more natural to expunge the second
‘in’ and accept the variant ‘unumquodque’ instead of ‘unoquodque’ (‘unumquodque indi-
viduorum hominis’ in the passage immediately after the one quoted).
Walter of Mortagne and the Identity Theory 159
Note that, as rational animals, Socrates and any human individual are
one and the same thing (idem). I do not say that they are essentially the
same thing, since according to that status (of being a rational animal)
and according to any other they are so opposed in their being that no
one is equal to any other, nor can be the same; even so they are identical
(idem), that is, non-diversified according to the status of man.24
Garland could find no better phrasing for the paradox of identity, a possible
way out of which is to distinguish between two relations: that of identity and
that of non-diversification. We must first understand how the relation of non-
diversification could be defined; that is, what definitions allow us to get a sense
for what Walter says, while remaining as close to the text as possible. I see four
possibilities (‘≈’ means ‘is non-diversified from’):
24 De Gen. Stat., Dijs, 104, 30; Romano, 103, 34: ‘Attende quod Socrates et unumquodque indi-
viduorum hominis in eo quod unumquodque est animal rationale mortale, sunt unum et
idem; non dico idem essentialiter, quia et secundum hunc statum et secundum quemli-
bet adeo opposita sunt in esse suo quod nullum eorum est aliquid aliorum, nec etiam esse
potest; sed sunt idem, id est indifferens, secundum statum hominis.’
25 For an introductory analysis and bibliographic information, see the article ‘The Identity
of Indiscernibles’ by P. Forrest for the Stanford Encyclopedia of Philosophy.
160 chapter 6
with a certain predication tree, where genera are divided into complementary
species. Given these limitations, it is questionable whether a more popular
similarity relation can meet Walter’s requirements. I cannot see how a relation
between tokens, or even a mild resemblance, may be a good candidate for the
relationship of non-diversification which exists, for example, between a tree
and my soul (which are both substances and therefore non-diversified at the
level of the terminal node of Porphyry’s tree).
The patterns that are relevant to Walter are the ones that allow the catego-
rization of objects (interpreted as properties or a statuses) under a node of
Porphyry’s tree. The following is an indexed part of an interpreted predication
tree:
…. …… …
Given two individuals one can say that they are i-equivalents, or non-i-
diversified, if and only if for each status/property f of level i, both have f or nei-
ther has f. For example, a man is non-i-diversified from another man that has
the same status/properties of level i. Non-i-diversified things form equivalence
classes for which it makes sense to choose a representative. Take the level 2
property of being an animal and choose one, for example Socrates, to repre-
sent the non-diversification in being animal (or more precisely, the class of
the partition induced by the relation). There will be things in the relation of
non-2–diversification that satisfy this property, and thus it can be said that
each of these is equivalent to the others ‘up to’ individual differences.26
26 It may be of some interest to recall what ‘identity up to isomorphism’ means in math-
ematics. The following friendly explanation is taken from Wikipedia: ‘In a mathematical
context, the expression ‘up to’ indicates that members of the same class of equivalence
Walter of Mortagne and the Identity Theory 161
Socrates Plato .…. Cicero
are considered a single entity. This is done because […] the differences in behavior be-
tween elements of the same class are negligible, and there are only differences between
different classes.’
162 chapter 6
Having stated his thesis, Walter seems to feel that some explanation is in order,
and begins by providing a series of questions (introduced by an item or item
notandum) and answers to possible objections (introduced by an item opponi-
tur). We shall follow the thread of this exposition by introducing the problem
(oi=objection), then the explanations given by the author (ai = answer).
o1 Boethius and other authors talk about the genus as the matter of things
partaking in it. If the genus coincides with the individual, how can this be con-
sidered the matter of something?
a1 Walter distinguishes between being essentially the matter of a thing and
being matter by non-diversification: being the matter of x, in the latter sense,
means nothing other than being non-diversified from x.
27 On the categorial status of the forms-properties in Priscian’s and Abelard’s commenta-
tors, see R. Pinzani, The Logical Grammar cit., 2003, 46–48, 54–56.
Walter of Mortagne and the Identity Theory 163
o2 In what sense can Socrates (as species) be said to be the matter of himself
(as an individual)?
a2 Walter recalls the example of the statue; the genus is non-diversified mat-
ter just as bronze before receiving the form that the artist intends to give it. It
must be understood that we have here an obscure dogma and a flawed inter-
pretation. According to the former, the universal is simultaneously present as
matter in many things; the latter concerns particular matter (bronze) present
only in a particular object (this statue). Having elaborated upon this theory,
Walter would not need to exploit the metaphor of the statue. On the other
hand, being a product of his time, he is not only a theorist but also a commen-
tator. As such, he must provide some additional explanation: ensuring that the
authorities support one’s own theory is the way this particular game is played.
The interpretation of the dogma is somewhat forced. The universal coincides
with the individual, and therefore, to say that the universal is essentially the
matter of an individual cannot mean anything other than that the individual
is its own matter. However, the meaning of this statement is not immediately
clear even for those who agree with the thesis of identity. Walter initially says
that ‘matter’ is meant metaphorically (per quondam similitudinem). He then
attempts to clarify what he means by introducing the example of paternal
seed, which can be considered the matter of the individual since it has ‘the
ability to receive the forms of man, i.e. rationality and mortality, before that of
Socrates.’ This second metaphor (if it is a metaphor), is not of much help: it is
difficult to think that as a man Socrates relates to himself in the same way as
the paternal seed relates to the individual which originates from it.
Walter persists in his attempt to elucidate the difference between Socrates
and Socrates, arguing that Socrates, according to the status of man, is to be
considered ‘more worthy’ and ‘prior.’ The relationship of precedence or (onto)
logical priority between species and individuals is dealt with in the literature,
but in this context the discussion seems rather extravagant. It is no coinci-
dence, then, that Abelard’s criticism of the theory of identity focuses precisely
on these interpretations of the traditional doctrine.
There is an important aspect of the metaphor of the priority of species.
According to the standard theory shared by most medieval logicians, sentenc-
es in the present tense are false if the subject denotes things that no longer
exist. If we say that Socrates coincides with the species once Socrates is dead,
statements about /man/ become false, contrary to the intuition and tradition
which consider them to be necessarily true. However, neither the theory of
the homunculus nor that of the precedence of the representative provide an
adequate explanation; what is actually needed is a careful logical analysis of
statements about abstract objects.
164 chapter 6
The noun ‘man’ here refers only to individuals who are (in actu sunt).
[… ] If one were considering the same noun according to the process of
imposition, perhaps it would not be predicated only of things that are,
or only of those that are not, as ‘man’ determines a mental act about the
rational mortal animal, whether such a thing exists or not; but that is
another story.28
From the point of view of the debate on universals, such a stance does not
change anything, but it makes clear the limits of the identity theory. In some
contexts – such as that of the definition – ‘man’ seems to denote something
like the ‘simple man’ or the rational mortal animal. The problem is that Walter
had labelled the latter analysis as a fairy tale well told. One ought to distinguish
between Walter’s position on universals, understood strictly in the Porphyrian-
Boethian sense, and the question of the meaning of nominal expressions. The
28
De Gen. Stat., Dijs, 107, 35; Romano, 107, 44: ‘Vox illa quae est ‘homo’ hic agit de individuis
tantum quae actu sunt; […] si acciperetur vox illa quae est ‘homo’ secundum propriam
inventionem, forsitan nec predicare ea tantum quae sunt, nec ea tantum quae non sunt,
cum ‘homo’ generet intellectum de animali rationali mortali aequaliter, sive sit, sive non
sit; sed haec alias.’
Walter of Mortagne and the Identity Theory 165
theory developed by Walter, as well as the one by Abelard, requires the postula-
tion of abstract entities such as statuses.
We have thus returned to the problem of the previous answer: when we
speak of man in general, as in the definition, we do not want to postulate any-
thing sensitive to changes in the extension of the terms. ‘Man’ refers to the
rational animal, whether or not rational animals exist. In the text just quoted
there is a ‘forsitan’ and a ‘haec alias’, which indicate that Walter reserves the
right to provide further explanations of this point. There are in fact at least
two possibilities: either we find an object, such as a status, which corresponds
to the meaning of the nouns used non-distributively, or we provide an analy-
sis in terms of quantified statements. Walter does not express himself clearly,
but what he says about fairy tales leads us to think that he was more inclined
toward an analysis of generic expressions in terms of quantified sentences, de-
spite the instability of their extension.
o4 If the individual and the species coincide, there are no ‘lower’ and ‘high-
er’ things. The objection evidently shows that the solution proposed by the
identity theorists is unable to give an account of what traditional sources say
about the relation of predication.
a4 This response ignores the objection: Walter simply states that ‘to be high-
er,’ for a certain individual thing, means to be considered according to a certain
status with respect to itself or to other individuals, while ‘to be lower’ means to
be considered in itself .
o5 As a species, Socrates is predicated of many things and is the matter of
several things. This objection is still based upon the common doctrine.
a5 Again, Walter offers no answers, but only a clarification of what he means
by ‘matter’: Socrates is essentially the matter of himself and of other non-
diversified things. ‘To be predicated of many things’ cannot mean anything
other than to be non-diversified from those same things.
o6 Boethius states that genera are and are not, meaning that they do not
exist in the same way that their individuals do.
a6 According to Walter ‘genera are’ means the same as ‘the individual ac-
cording to the status of genus exists’. The statement ‘genera are not’ can be
understood in two different ways: as a simple denial of the first statement, in
which case it is false, or in the sense that the individual has other properties in
addition to the status of genus, in which case it is true.
o7 The sentence ‘every man is an animal’ is ambiguous, since the subject
refers to different individuals.
a7 Walter’s answer is not one of exemplary clarity: single men are referred to
by the subject, but since they are not considered according to their individual
status but according to the status of the species, there is no ambiguity. The fact
166 chapter 6
that Walter does not have any (proto) theory of quantification at his disposal
may perhaps serve as partial justification for such a hasty assertion.
o8 In the syllogism ‘every man is an animal, Socrates is a man, therefore
Socrates is an animal,’ the middle term stands for any man insofar as he is a
man in determinate way (‘determinate’), while in the second premise the same
things are not predicated in the same way; the syllogism therefore does not
work. A real critic would certainly have in mind some technical use for ‘to be f
in a determinate way’. In explicating the thesis of the simple man criticized by
Walter, we noticed some terminological anticipation of the theory of supposi-
tio. In this case too, the distinction between ‘determined’ and ‘undetermined’
seem to suggest some knowledge of the contemporary debate on contextual
changes of meaning.
a8 Walter’s answer to this objection is just as disappointing as the previous
one: both premises refer to men according to their status as men, and therefore
there is no equivocation.
o9 According to the status of animal, Socrates is predicated of several things,
according to the status of Socrates he is predicated of one only, thus it is the
same thing, and not several, that is predicated.
a9 Again, Walter rephrases several standard approaches: ‘Socrates is predi-
cated of several things according to the status of animal’ means ‘several things
are similar to Socrates in the sense that they are men’. When we say ‘accord-
ing to the status of Socrates, Socrates is predicated of one thing’ we mean ‘it
is not true that several things are similar to Socrates in the sense that they
are Socrates.’ According to this rephrasing, the statements in o9 do not contra-
dict each other. Walter’s logical-philosophical analysis may appear to be too
nonchalant; however – at least in this case – he is in good company. To some
extent, the non-realist interpretations of the traditional theory involve some
forcing of Porphyry’s and Boethius’ theses. In Abelard’s Logica Ingredientibus,
for example, authoritative theses of a realist bent are systematically interpret-
ed as if they were concerned with linguistic expressions and their meanings.29
o10 According to Porphyry, the genus differs from the individual in that the
former is predicated of several things, while the latter is predicated only of
one. This difficulty is different from that in o9: here it is not a question of un-
derstanding how the same individual is predicated both of several things and
29 The formula ‘many things are alike..’ is found literally identical in Abelard. Boethius too
speaks of communion of several things of the same kind, for example: ‘ut enim plura
animalia quoniam habent quiddam simile, eadem tamen non sunt, idcirco eorum genera
perquiruntur, ita quoque quoniam genus quod in pluribus est atque ideo multiplex habet
sui similitudinem quod genus est’ (In Isag. II, 162, 7–10).
Walter of Mortagne and the Identity Theory 167
30
De Gen. Stat., Dijs, 111, 48; Romano, 111, 53: ‘Socrates secundum statum animalis differt a se
ipso secundum statum Socratis non per illam predicationem quae est in eo essentialiter,
168 chapter 6
We have seen how necessary it is, for the purposes of the identity theory, to
postulate relations of i-non-diversification. These are relations of equivalence,
which means that they are characterized by reflexive, symmetric and transitive
properties. Walter is now telling us that there are two ways to be in a relation
of non-diversification, essentially and non-essentially, the latter being, so to
speak, ‘simple’. This seems to entail a distinction between 0–relation and i≠0–
relations of non-diversification. Consider the following diagram, in which the
dotted arrows show the sub-relation of non-essential-non-diversification, and
the continuous ones the sub-relation of essential-non-diversification (which,
in fact, coincides with identity in the strict sense):31
Socrate Platone
The sentence ‘Socrates differs from himself not because of the predication
which is essentially in him, but because of the one which he has by non-
diversification’ can be read as follows: the genus, that is, the individual according
to the status of genus, is essentially predicated of itself; the same individual, ac-
cording to the status of individual, is predicated essentially of itself. Therefore,
in regard to essential predication, the individual as genus does not differ
from itself as individual. The individual, however, differs from itself by non-
essential-non-diversification. Indeed, the sentences ‘Socrates is predicated of
something different from himself’ and ‘Socrates is not predicated of anything
but himself’ are both true secundum quid. The former is true of Socrates con-
sidered as species, the latter is true of Socrates considered as such.
Where have all these considerations brought us? Porphyry’s position,
which Walter is trying to preserve, is that the genus is predicated of several
things, while the individual is predicated of one thing. Walter is able to ex-
plain how Socrates as genus is predicated of several things (by non-essential-
non-diversification) and at the same time be predicated of one thing only (by
essential-non-diversification). In other words, we cannot distinguish between
two things such that one has a property and the other does not. Rather, we can
sed per illam quam habet per indifferentiam, scilicet per illam quam habet Plato secun-
dum statum animalis essentialiter’.
31 See the interpretation supplied above. Given two elements of the level i, one can say that
they are i-equivalents, or non-i-diversified, if and only if for each status/property f of level
i both have f or neither has f; so at level 0 only Socrates is 0–equivalent to himself, since
nobody has all the same properties Socrates has.
Walter of Mortagne and the Identity Theory 169
at most say that the same individual has two different properties, just like the
Boethian line observed from two different perspectives. After all these contor-
tions, we do not seem to have strayed far from a10b.
a10d Walter does not seem satisfied with the previous solution and proposes
a different one. The property of being predicated of several things can be at-
tributed to both the genus and the individual, and yet, ‘quod mirum est’, the
genus differs from the individual because of the same property. The explana-
tion, then, is that Socrates as genus is predicated and as individual is not. In
other words, the same property is ‘removed’ from Socrates as an individual.
This can be seen in the same way as contrary qualities: white is contrary to
black, but a white body is not the opposite of the same body which has be-
come black. To understand this explanation we must establish a connection
between Socrates as a species and the color white on the one hand, and, on the
other, between Socrates as an individual (the supposed foundation of Socrates
as a species), and what ‘supports’ whiteness, i.e. the white body. In the former
there would be a certain property that is not in the latter. The variant explana-
tions do not ultimately seem very different; Walter tries to convince us that a
certain property and its opposite can be found in one and the same thing seen
from different points of view.
o11 If Socrates is destroyed as genus, the rationality that ‘resides in’ the genus
/animal/ is destroyed. Conversely, if rationality is destroyed, Socrates could not
survive in his individual status as rational being, but he could survive in his
generic status.
a11 Walter asks what it means to destroy a thing according to a certain
status. Obviously the fact that Socrates is destroyed does not entail that the
rationality which is in Plato is destroyed, but if one destroys Socrates ‘by non-
diversification’, rationality everywhere is destroyed. This solution requires
agreement on the meaning of ‘destruction by (non-)diversification’; by de-
stroying rationality ‘by diversification’, not only is Socrates destroyed, but also
any bearer of rationality. In this case, with all bearers of rationality destroyed,
rationality itself ceases to be, or in other words, is no longer instantiated.
The objections and replies end here. They are mostly intended to clarify the
meaning and scope of Walter’s position. In fact his theory requires several
qualifications to be understood correctly. The aim is to justify the identity the-
sis while trying to save the apparently divergent authoritative opinions. It is
not entirely clear whether these criticisms had been foreseen by Walter before
they were actually made; the fact is that Abelard, as far as we know the main
opponent of the identity theorists, focuses his attacks on the points considered
above.
chapter 7
The De Generibus et speciebus1 can be divided into two parts: the first deals
with the notion of whole, the second deals with the issue of universals. The
opening paragraphs, on the concepts of (continuous and scattered) whole in-
troduce the technical equipment needed to address the following discussions;
or, from another point of view, contain scholastic material not directly linked
to the following treatise. Scholars have recently raised doubts about the unity
of the work and its attribution, in particular regarding the connection between
the first and the second part; actually, according to A. Arlig, within the first part
one should distinguish between a collection of school notes on the Topic and
a treatise on composition and mereological change.2 In this work we are in-
teressed in the second part which contains the criticism to the contemporary
theories on universals, the presentation of the collectio theory and the author’s
solution to the problem of universals.
1 The De Generibus et speciebus (= De Gen.), has been published by V. Cousin and inserted
into the unpublished Abelardian works: V. Cousin, Ouvrages inédites d’Abélard, pour servir à
l’histoire de la philosophie scolastique en France, Paris 1836. The edition is based on the manu-
script Paris, Bibiothèque nationale, lat. 11168 (p. 168ra–179va). The text of De Generibus is also
preserved in the manuscript Orléans, Bibliothèque municipal, 266 (p. 153a–166a). P. King
has re-edited the text in ‘Pseudo-Joscelin: Treatise on Genera and Species’, Oxford Studies in
Medieval Philosophy, 2 (2014) 132–211. The references are to the King’s edition.
2 According to A. Arlig – in partial discordance with the previous conclusions of his P.D. dis-
sertation – the De Generibus ‘is not a unified treatise. In particular.. the first part of the trea-
tise (§ 1–20) … is only superficially related to the second part (§ 21–31) … this second part
shares many of the same metaphysical commitment found in the third and largest part of De
Generibus (§ 32–204) … it is not implausible that the second and third parts are by the same
author’ (‘Early medieval solutions to some mereological puzzles: the content and unity of the
Generibus et Speciebus’, in Arts du language et théologie aux confins des Xie / XIIe siècles, ed.
I. Rosier-Catach, Brepols, Turnhoult 2011, 485–508; 485–86).
3 Joscelin of Soissons or Goslenus was a master at the cathedral school in Paris from 1110 and
later bishop of Soissons. John says: ‘est et alius qui cum Goslenus Suesioniensi Episcopo,
universalitatem rebus in unum collectis attribuit, et singulis eandem demit’ (Metalogicon, ed.
G.B. Hall, Turnhout, Brepols, 1991, lib.III, cap. 17, pp. 89–93). This author therefore appears to
support a theory similar to the one criticized by Abelard. However, John speaks of Joscelin’
position as one that attaches universality to the collection of things. The Logica Ingredientibus
was written around 1120 (cf., also for bibliographical references J.E. Brower – K. Guilfoy,
Introduction, in J.E. Brower – K. Guilfoy (edd), The Cambridge Companion to Abelard,
Cambridge, Cambridge UP 2004, p. 7).
4 This attribution has to be considered in any case according to the naming convention I state
above (see Introduction).
172 Chapter 7
if Joscelin knows Abelard, why does he not discuss him extensively? An argu-
ment in favor of the priority of De Generibus over the Logica Ingredientibus
is as follows: Abelard speaks of the theory of collectio as a kind of theory of
non-difference. In principle there is no reason why a proponent of a theory of
collectio should also embrace a form of non-difference: one can simply main-
tain that things constituting a certain collection have a common property,
without claiming that they are non-diversified as to that property. Joscelin, on
the other hand, actually does talk about the non-diversification of the essences
constituting a universal collection. Conversely, the following circumstances
support the priority of the Ingredientibus over De Generibus: a) The theory
of collectio criticized by Abelard is a primitive or ‘naïve’ version compared to
the one put forth in De Generibus; b) Joscelin knows contemporary semantic
threads and makes some use of them to argue in favor of his solution. In par-
ticular, he speaks of translatio terminorum, which could be seen as a more up-
dated version of Abelard’s semantic theory.
It may have been the case that Abelard misunderstood the theory presented
by Joscelin or did not consider it carefully enough, but it seems rather strange
that his criticisms are the very same as those to which Joscelin provides an-
swers. On the other hand, Joscelin’ silence on Abelard’s position5 could be
explained by the fact that he does not want to antagonize a famous and contro-
versial character. All things being equal, if one must choose who read whom, it
seems slightly more likely that Joscelin read Abelard.
5 Apart from the reference he makes to a semantic interpretation of universals when exposing
a nominalist thesis in Roscelin style (universals as voces): cf. De Gen., 74 (153) ff.
6 Cf. De Gen., 32 (132–34).
The De Generibus et Speciebus and the Theory of Collectio 173
(b) is presented by Joscelin as the genus of c–d.7 The first thesis considered is
(d), the classic realist stance attributed to William of Champeaux. In Chapter 4
we considered Joscelin’ exposition of TR1, we now consider its critical remarks
to the main thesis of realism.8
c1. If the universal is found in several individuals at the same time, the same
thing will be simultaneously in Rome and Athens; in fact, the universal man is
present where Socrates is and also where Plato is.9
c2. The second difficulty concerns the presence of conflicting properties in
the same individual. For example, disease resides entirely in the genus /ani-
mal/, and therefore is in Socrates but also in Plato, who is not ill, but healthy.
The loophole that lets us out of the problem is that the animal in Socrates is
sick, but the animal itself is not sick. Joscelin, along with other critics of real-
ism, does not accept this escape route because he believes that an advocate of
TR1 must necessarily accept the identity of the generic animal and the animal
in Socrates. Neither can it be argued that the animal as universal is not sick, be-
cause in the same way one could say that it is not sick as a particular. Moreover,
if we say the animal is not sick in its universal status, we must clarify what we
are talking about, since the status does not seem to be anything different from
the animal itself.
c3. Rationality and irrationality, added to the genus, give rise to par-
ticular species. As in the case of illness, both opposites happen to be in the
genus /animal/. The first loophole considered by Joscelin is an adaptation
of Porphyry’s position: the differences are not present in the constitution
7 De Gen. 32 (134), 1–5: ‘Alii vero res generales et speciales universales et singulares esse dicunt;
sed et ipsi inter se dissentiunt. Quidam enim dicunt singularia individua esse et species et
genera subalterna et generalissima, alio et alio modo attenta. Alii vero quasdam essentias
universales fingunt quas in singulis individuis totas essentialiter esse credunt.’
8 We list below the criticisms to realism, the theory of non-difference and vocalism under the
abbreviations, ci. The remarks on Joscelin’ theory and the answers to these remarks will be
listed under the abbreviations, ri.
9 The example ‘piper venditur hic et Romae’ discussed in the treatises on the properties of terms
contains a similar difficulty; see for example: l’Index Sophismatum et exemplorum in the ap-
pendix to L.M. De Rijk, Logica Modernorum, Vol. II, Part 2, Van Gorcum, Assen 1967; William
of Sherwood, Introductiones in Logicam, ed. M. Grabmann, Sitzungsberichte der Bayerischen
Akademie der Wissenschaften. Philosophisc – historische Abteilung, 1937, Heft 10, Munchen
1937, 77, 28 – 78, 2; Ockham Summa Logica, ed. P. Boehmer – G. Gál – S. Brown, Editiones
Istituti Franciscani Universitatis S. Bonaventurae, St. Bonaventure N.Y., 1974, 205, 143–155.
Some considerations on this subject can be found in H. Parret, History of linguistic thought
and contemporary linguistics, de Gruyter, Berlin 1976, 214.
174 Chapter 7
(in actu constitutionis) of the genus, i.e. they do not enter into the definition of
its essence.10 So far, there can be no objection. The problem raised by Joscelin
and other critics of realism is that Porphyry’s thesis itself (that the differences
are in potency in the genus, or as it is said here, do not determine its constitu-
tion), together with TR1 leads to the difficulty of having differences that are at
the same time in act and potency.
Other defenders of TR1 say the differences supervene upon the genus, but
are not grounded in it. Joscelin objects that the genus sustains the forms, and
therefore differences are grounded in the genus. This answer is rather strange;
it is not clear why those who interpret the presence of the differences in the
genus as a form of supervenience – whatever this may mean – should accept
the latter conclusion.
After this criticism of realism’s main thesis, Joscelin goes on to consider
the theory of non-difference. Let us begin by seeing how the thesis of non-
difference is presented:
Nothing exists except the individual; but singular things, considered from
different points of view, are species, (subordinate) genus, and highest
genus. Thus Socrates, considered according to the nature by which he is
subject to the senses (and according to the nature that ‘Socrates’ signi-
fies with respect to him), is an individual, for he has a property which
is not present in anything else. From another point of view, he is a man
[…]. We sometimes have a concept that is not about that to which the
expression ‘Socrates’ refers. Rather, putting Socrates’ property aside, what
is conceived about Socrates is only what the expression ‘man’ signifies
[…] and in this sense (Socrates) is species. In fact, it is essentially predi-
cable of several things in accordance with the same status. If the concept
disregards rationality and mortality, and only applies to what ‘animal’
signifies, in this status Socrates is genus. If, leaving out all the forms, we
consider Socrates as what connotes the substance he is highest genus.11
10 Porphyry argues that the differences are potentially in the genus, so in principle the genus
without differences should remain in its current actual state of genus. Realists, however,
argue that the genus is present in the individual. On the other hand, a human individual
is actually a rational animal, so rationality is present in the animal that is actually present
in the individual. If we remove rationality, the individual dies, but according to Aristotle
the second substance ceases to be along with individual instances.
11 De Gen. 50 (142), 30 – (144), 12: ‘Nihil omnino est praeter individuum; sed et illud aliter et
aliter attentum specie et genus et generalissimum est. Itaque Socrates in ea natura in qua
subjectus est sensibus, scilicet secundum illam naturam quam significat de eo ‘Socrates’,
individuum est, ideo quia tale est proprietas eius numquam tota reperitur in alio. Est
The De Generibus et Speciebus and the Theory of Collectio 175
enim alter homo […] quandoque habetur intellectus non concipiens quicquid notat haec
vox Socrates, sed, Socratitatis oblitus, id tantum perspici de Socrate quod notat idem
‘homo’ […] et secundum hanc attentionem species est, est enim praedicabilis de pluribus
in quid de eodem statu. Si intellectus postponat rationalitatem et mortalitatem, et id tan-
tum sibi subiciat quod notat haec vox ‘animal’, in hoc statu genus est. [Quod] si, relictis
omnibus formis, in hoc tantum consideremus Socratem quod notat [inde] substantia,
generalissimum est.’
12 Ibid.: ‘Si quis dicat proprietatem Socratis in eo quod est homo non magis esse in pluriuso
quam ejusdem Socratis in quantum est Socrates, aeque enim homo qui est Socratis in
nullo alio est nisi in Socrate, sicut ipse Socrates, verum quidem concedunt; ita tamen
determinandum putant: Socrates in quantum est Socrates nullum prorsus indifferens
habet quod in alio inveniatur; sed in quantum est homo plura habet indifferentia quae in
Platone et in aliis inveniuntur.’
13 In presenting the criticisms of the non-difference (identity) theory we will not use the ter-
minology introduced in the previous chapter. We will maintain the terms ‘non-different’,
‘different’, ‘difference’ etc., since the distinction between the different senses in which two
things can be said identical or (in)different is not relevant for a number of criticisms.
176 Chapter 7
In the following pars destruens, Joscelin does not mention individual prop-
erties, but prefers to focus his criticism on the disagreement between the the-
ory of non-difference and the Boethian commentaries on Porphyry. His critical
remarks on the non-difference (or identity) theory are summarized below.
c4. Porphyry says that there are infinite individuals while there are ten high-
est genera. From this it follows that individuals cannot be highest genera. The
supporters of the non-difference theory reply that there are ten highest gen-
era by non-difference, but they are essentially infinite; for example, Socrates,
Plato, and other individuals coincide with /substance/, while only one of the
individuals of the same genus is taken as representative of the relationship of
non-difference.
c5. The genus is inherent (inhaeret)14 to the species, while the lowest species
is inherent to the individual. If the individual is species (genus), it must do
two jobs. The reply of identity theorists is the same one we encountered in the
Walter’s text on non-difference: Socrates is not essentially inherent in anything
other than himself, but by non-difference is inherent in many individuals.
c6. The species is a conceptual collection (cogitatio collecta) deriving from
the consideration of the similarity of many individuals, while the genus is a
collection that comes from the consideration of several species. Therefore,
species-genera do not coincide with individuals. According to Joscelin identity
theorists reply that the individual ‘collects’ (colligitur) itself. Walter of Mortagne
does not say anything like that; here he may have simply observed that the
formula ‘the species is obtained from a number of things having a property in
common’ means ‘individual x has a property in common with others.’
c7. The universal is predicated, and the particular is that of which the uni-
versal is predicated. If the universal is the particular, we cannot distinguish be-
tween subject and predicate. Joscelin does not report any counter-argument,
but Walter would have something to say about it.
c8. If Socrates is universal and the universal is not particular, it follows that
Socrates is not Socrates. Identity theorists make their typical move of reinter-
preting the texts somewhat metaphorically. Joscelin is annoyed by this ap-
proach (videte quante impudentie sunt!), though he has no reason to be, given
that he shares the same attitude as his opponents.
c9. The universal is different from the particular in that the former is predi-
cated of the latter, but not vice versa. According to Joscelin the objection is
decisive, but identity theorists do not give up, and continue to argue that no
singular thing, as singular, is universal. This means that from their point of
14 In this context Joscelin does not seem to attribute to ‘inherit’ a different meaning from ‘is
predicated’.
The De Generibus et Speciebus and the Theory of Collectio 177
As these authors do not find a natural way to interpret what the authori-
ties say so clearly, they either say that they are lying, or, in trying to find
an explanation, being unable to remove the skin completely [in the sense
of clarifying the deeper meaning of the text], they only nick the surface.16
c12. Linguistic expressions are ‘ephemeral’ in both phonic and written form,
though in different ways. If a philosopher is a nominalist, in the sense that he
asserts that only concrete entities are located in a certain place at a certain
moment, then – Joscelin says – phonic expressions are not among these things.
15 This argument corresponds to Walter’s point at r10 (cf. the chapter on Walter of Mortagne’s
theory of identity).
16 De Gen. 79 (156), 1–3: ‘His autem tam apertis auctoritatibus rationabiliter obviare non va-
lentes, aut dicunt auctoritates mentiri, aut exponere laborantes, quia excoriare nesciunt,
pellem incidunt.’
178 Chapter 7
If, instead, a nominalist posits that expressions are not located in (a phonic)
space, he is compelled to assume abstract types, in spite of his attempts at
metaphysical rigor.
c13. Usually species are considered to be composed of genus and difference;
if genera and species are expressions, it is unclear what it means that an ex-
pression is made up of another expression. Joscelin reports a possible reply of
one of Roscellinus’ followers: ‘when you say genus is the matter of the species
you use a metaphor, what you actually want to say is that the meaning of genus
is the ‘matter’ of the meaning of ‘species’.’ This shift from the expressions to
their meanings should catch the reader’s attention, since Abelard adopts the
same strategy. The criticism that we find here is simple but effective: Joscelin
notes that accepting universal meanings is incompatible with a metaphysical
position according to which there are only individuals. As we shall see, Abelard
has something to say about this, but in view of this kind of attack, he is decid-
edly on the defensive.
c14. The question arises again about the sense in which the term ‘animal’
may be the matter of the term ‘man’. In exposing of this point, the concept of
non-difference is recalled with no apparent reason. From a historical point of
view, it is useful to compare what Joscelin says in this regard with Garland the
Computist’s commentaries on Isagoge and Categories.
The pars construens of the treatise begins with a definite metaphysical stance:
17 De Gen. 87 (158), 5–10: ‘Unumquodque individuum ex materia et forma compositum
est, ut Socrates ex homine materia et Socratitate forma […] et sicut Socratitas, quae for-
maliter constituit Socratem, nusquam est extra Socrate, sic illa hominis essentia, quae
Socratitatem sustinet in Socrate, nusquam est nisi in Socrate.’
The De Generibus et Speciebus and the Theory of Collectio 179
As the reader may recall, in the metaphysical landscape of TR2 and TR3 there
exist variants of the main thesis according to which there are discrete non-
sharable forms (essences). Apparently, Joscelin is saying something similar;
however, he quickly reaches a point of divergence from supporters of TR2–3:
I say that the species is not just that essence of man that is in Socrates or
in another individual, but the collection constituted by their individual
matters.18
‘Things of that nature’ might refer to individual things having an essence, and
not to the essences themselves. In what follows, however, Joscelin continues
to talk about collections of essences consisting only of a generic matter and
specific differences:
By saying that the individual essence of man bears (sustinet) Socrates’ forms,
Joscelin implicitly distinguishes between the bearer (= specific matter) and
the thing considered together with all its properties. Thus, the essences con-
stituting universals do not seem to coincide with individuals. Rather, it is what
remains once individual things have been stripped of each sub-specific or sub-
generic feature.
18 De Gen. 87 (158), 10–12: ‘Speciem igitur dico esse non illam essentiam hominis solum quae
est in Socrate, vel quae est in aliquo alio individuorum, sed totam illam collectionem ex
singulis illis materiis factam.’
19 Ibid., 17–19: ‘Unaquaeque essentia hujus collectionis quae humanitas appellatur, ex ma-
teria et forma constat, scilicet ex animali materia, forma autem non una, sed pluribus,
rationalitate et mortalitate et bipedalitate.’
20 Ibid., 87 (158), 24 – 89 (158), 33 – 89 (160), 1: ‘Hanc itaque multitudinem essentiarum ani-
malis quae singularum specierum animalis formas sustinet, genus appellandum esse dico
[…] singulae essentiae illius multitudinis quae animal genus dicitur, ex materia aliqua
essentia corporis et formis substantialibus, animatione et sensibilitate, constat.’
180 Chapter 7
a) For Joscelin universals are not Platonic entities. If species and genera are
interpreted as sets, at most they can be said to be abstract. As we shall see,
however, there are reasons to think that this is not the case.
b) The relationship between universals and particulars is a part-whole rela-
tionship, in the sense that the particular matter of the individual is part
of a universal.
For these reasons, Joscelin cannot be placed among the supporters of a realist
thesis in the style of William of Champeaux and his school. We can speak of
realism, however, in many senses, and some authors such as Abelard believe
that positing real universals is sufficient to be labeled as a realist. One might
be tempted to go further and argue that the genre of realism at hand has no di-
rect connection with the principles of predication and Porphyry’ problem. Let
me explain the point: suppose that one wants to consider the genus /man/ as
an aggregate of human parts. The instance, then, of the predication principle
‘Socrates is a part of the human aggregate iff Socrates is a man’ should be valid.
This statement does not, however, contain a special occurrence of ‘is a part
of’; rather, the same expression has a meaning equivalent to the one it has in
the (false) statement ‘the roof is a part of the house aggregate iff the roof is a
house.’
One who offers an interpretation of universals as continuous essences (such
as a house) should, therefore, disregard the principle of predication. In doing
so, he distances himself from the problematic picture drawn in the first chap-
ter. On the other hand, I find it unlikely that the example of the house, repeat-
edly cited by Joscelin, has paradigmatic value. The easiest way to think of the
parts of the universal /man/ is to consider them as sheep in a flock, or some
men among the people, that is, as ‘multitudes with confined parts,’ to quote
Nicomachus-Boethius. All is not lost, then, for the principle of predication, as
demonstrated by the following example:
The De Generibus et Speciebus and the Theory of Collectio 181
The principle is applicable to totalities only when they are named in a certain
way; for example, the totality of the trees (not the forest), or the totality of men
(not the people). A passage later in the text serves as confirmation that Joscelin
is thinking of multitudes with confined parts:
21 De Gen. 87 (158), 14–24: ‘Quae tota collectio, quamvis essentialiter multa sit, ab auctorita-
tibus tamen una species, unum universale, una natura appellatur, sicut populus, quamvis
ex multis personis collectus sit, unus dicitur […]illud animal quod formam humanitatis
quae in me est, substinet, illud essentialiter alibi non est, sed illi indifferens est in singulis
materiis singulorum individuorum animalis.’
182 Chapter 7
The man who bears Socrates-ness is not essentially the same as the one
who bears Plato-ness.22
I call genus the multitude of animal essences that bears the forms of
the individual species of animal.23
Our position is that some essences of the animal are informed by ratio-
nality to constitute /man/; others by irrationality to constitute /donkey/.24
A substantial item is formed by the ultimate species and its own form;
the species by genus and difference.25
When read on their own, these texts lead one to think of properties which
are unique to the essential parts of species/genus. Upon further inspection,
however, one finds circumstantial evidence that properties are as individual as
generic-specific essences are:
One must consider the fact that the individual essences of the multitude
called /animal/ genus are made of a certain material, the essence of the
body, and of essential forms, i.e. animation and capacity to perceive,
forms which, as has been said of the animal, are not essentially elsewhere.26
That part of humanity (illud humanitatis) which is in me […] is one of
the essences […] which make up the human species […] similarly, that
part of rationality (illud rationalitatis) […] is one of the essences which
make up the rationality which is the difference.27
22 Ibid., 20–21: ‘illud hominis quod substinet Socratitatem, illud [non] essentialiter <non>
[= lectio V. Cousin] sustinet Platonitatem’;
23 Ibid., 24–25: ‘Hanc itaque multitudinem essentiarum animalis quae singularum
spiecierum animalis formas sustinet, genus appellandum esse dico.’
24 De Gen. 136 (178), 7–10: ‘Hoc enim habet nostra sententia quod pars essentiarum animalis
quae illud genus faciunt, informatur rationalitate ad facendum hominem; pars vero ir-
rationalitate ad facendum asinum.’
25 De Gen. 148 (184), 13–14: ‘Individuum substantiae ex specialissima specie et forma propria,
species vero ex genere et forma differentia.’
26 De Gen. 89 (158), 32 – (160), 1: ‘sciendum est quod singulae essentiae illius multitudinis
quae animal genus dicitur, ex materia aliqua essentia corporis et formis substantialibus,
animatione et sensibilitate, constant, quae, sicut de animali dictum est, nusquam alibi
essentialiter sunt.’
27 De Gen. 196 (206), 10–14: ‘Illud humanitatis quod in me est, non est vel genus vel species
vel individuum humanitatis, sed una tantum de essentiis ex quibus colligitur humanitas
quae est species. Eodem modo illud rationalitatis quod in me est, nil aliud est nisi una de
essentiis de quibus conficitur rationalitas quae differentia est.’
The De Generibus et Speciebus and the Theory of Collectio 183
We see, then, that Joscelin refuses to consider species and genera as abstract
and universal entities, and supposes that they are multiplicities of individual
essences. We cannot expect him, therefore, to have much sympathy for any
common entity. Joscelin’ tree of predication could be similar to the one below:
genus /animal/ = [substance1 + being animated1 + being animal1….
substance m + being animatedm+ + being animalm]
species /man/ = [substance1 + being animated1 + being animal1 + rationality1
+ mortality1…. substance9 + being animated9 + being animal9 + rationality9
+ mortality9]
1 2.…..………9
Assume that we are counting nine men. Individual one is identical to the in-
dividual essence (substance1 + being animated1 +being animal1 + rationality1
+ mortality1 + particular individual properties). The indices I have introduced
facilitate understanding, but also show exactly where the problem lies. Parts
of the species-genera are made up of essences and particular properties; but
upon what does the assigning of properties to individuals depend? In other
words, if a certain property f is particular, why must we talk about Socrates’
property f and Plato’s same f? What Joscelin says about the non-difference of
individual properties can be explained by looking at other similar theories that
are closer to us, but these explanations do not always prove to be suitable. We
can have some intuition about the fact that this red looks like this other red,
but the intuition weakens for properties which do not allow nuances (height,
having 3 dimensions, being equal .. ). Again, in the case of essences, we do not
even know exactly what it means to have an individual essence; all that we
actually have is a common noun.
184 Chapter 7
7.4 On Meaning
Just like ‘rational’ and ‘man’, any other universal expression signifies
something else, for the imposition of what it primarily means. I mean,
‘rational’ or ‘white’ was imposed on Socrates or other sentient things to
denote them because of their forms, namely rationality and whiteness,
which these expressions primarily mean. Similarly, ‘man’ was imposed
upon any individual materially constituted from <being> man, because
of its matter, that is, the species that it primarily means. 28
Abelard would most likely agree with this text,29 with one caveat. He would
point out that it is inappropriate to say that the main meaning is the species,
since the word ‘species’ only refers to linguistic expressions. We should say that
the main meaning of a general noun (as ‘man’) coincides with its denotation
28 De Gen. 101 (164), 5–11 : ‘Sicut rationale et homo, sic et quodlibet aliud universale substan-
tivum alterius nomen est, per impositionem quidem ejus quod principaliter significat.
Verbi gratia : rationale vel album impositum fuit Socrati vel alicui sensilium ad nominan-
dum propter formas, id est rationalitatem et albedinem, quas principaliter significant.
Eodem modo homo impositum fuit cuilibet materialiter constituto ex homine ad nomi-
nandum, propter eorum materiam, scilicet speciem quam principaliter significaret.’
29 See, inter alia, the famous pages of Dialectica 112–114.
The De Generibus et Speciebus and the Theory of Collectio 185
when it is used as a species name (for instance in the definition ‘man is the ra-
tional mortal animal’). Apart from this distinction (which, admittedly, does not
seem substantial), the meaning to which Joscelin refers differs from Abelard’s
in the sense that the former is a collective whole, while the latter is a status,
an entity of uncertain ontological collocation, otherwise characterized as ob-
ject of discourse, common nature, common cause of imposition, or plurality
of things.30
In discussing the meaning of nouns, Joscelin encounters the problem of
the contextual use of expressions. ‘Man’ in the context ‘every___ runs’ seems
to mean something different from what it means in the context: ‘___is a spe-
cies’ or ‘the definition of ___ is mortal rational animal’. The theory of suppositio
provides an explanation as to why a term can have different meanings. In the
first half of the twelfth century, we find the concept of translatio terminorum
(transfer of term meaning). As far as I know, this subject is mainly discussed
by Abelard and by Joscelin himself. In both his logical and theological works,
Abelard makes considerable use of the notion of translatio.31 Below we list the
main contextual changes in meaning he considers:
1. ‘Rident’, in the poetic context ‘prata rident’, means something like ‘are full
of beautiful flowers’.
2. The term ‘homo’ in ‘‘homo’ est vox’, is intended to mean ‘by translation, not
by some institution.’32
30 In a passage of the Logica Ingredientibus. Abelard says that the status of man (or esse
hominis) can be seen as things (i.e. men) statutae. This phrasing, which is not a very com-
mon one, is, in some ways closer to Joscelin’ way of explaining meaning.
31 In the commentary on the Categorie Boethius introduces the notion of translatio, in the
context of a discussion of ‘deliberate’ equivocation (equivocatio consilio: cf. Boethius, In
Cat. 166D–167A). J. Jolivet (Arts du language et théologie chez Abelard, Paris, Vrin, 1982)
lists possible terminological and doctrinal precedents: Cic., Top. VI, 31–32; De Orat. III,
36, 155; Quint., Inst. Orat. VIII, 6; Boethius, In Cic. Top. 1107B - 1108B; Donatus (for the defi-
nition of ‘trope’), Ars Grammatica 399. To these we must add the ones by Boethius and
De Gen.. 101–107. As for the problem, some indications are found in Jolivet (op. cit.) and
D.E. Luscombe, Peter Abelard, in Dronke, P. (ed.),. A History of Twelfth-Century Western
Philosophy, Cambridge, Cambridge U.P., 1988. Something more is said by L.M. de Rijk, in
Logica Modernorum: A Contribution to the History of Early Terminist Logic. I. On the Twelfth
Century Theories of Fallacy, Assen, Van Gorcum, 1962, pp. 49–61, Logica Modernorum II-1,
The Origin and Early Development of the Theory of Supposition, Assen, Van Gorcum, 1967,
p. 492 and ff..
32 Log. Ing. 336, 12–20; 378, 9–17; 399, 26–30.
186 Chapter 7
3. The generic and specific nouns ‘are transferred to their things, as one says
the species is made of genus and difference, i.e. the thing of the species is
made of the thing of the genus’.33
4. In ‘a man was sent by God’, the term ‘man’ refers to past men by a transfer
of meaning.34
5. The words of natural language used to talk about God violate their institu-
tion (institutionem excedere).35
6. Ordinary expressions are sometimes used in a technical sense.36
According to Abelard, the changes in meaning which occur ‘by necessity’, such
as when an expression already in use is given to something in order to signify
it, lead to equivocation. This seems to mean that all proto-suppositive transfers
of meaning involve an equivocation of terms. Joscelin observes:
Joscelin is talking about the fact that a term like ‘man’ has two meanings: what
it is imposed upon, and the main meaning, i.e. the species. In contexts such
as ‘every man runs’, what matters in determining the truth conditions is the
33 Log. Ing. 30, 18–20. Here the expressions ‘res speciei’ and ‘res generis’ suggest that Abelard
wishes to signal a transition from a nominal point of view to a real one.
34 Log. Ing. 350, 10–11.
35 Theologia Christiana, ed. E.M. Buytaert, (CCCM), Turnhout, Brepols, 1969, p. 245: ‘Unde
in Deo nullum propriam inventionem vocabulum servare videtur, sed omnia quae de Eo
dicuntur, translationibus et parabolicis aenigmatibus involuta sunt.’
36 Cf. inter alia Theol. Christ. 245: ‘Cuiuslibet artis tractatus propriis utitur verbis et unaquae-
que doctrina propriis locutionibus gaudet et saepe eiusdem artis tractatores verba variare
delectat.’ Observation that has many confirmations in the logical works, often connected
with the theme of translation.
37 De Gen. 107 (166), 8 – 111 (166), 29: ‘Vocabula quae imposita sunt rebus propter aliud signifi-
candum, id est principaliter circa ea, quandoque transferuntur ad agendum de principali
significatione […] non est autem aequivocatio […] concedimus itaque hanc translatio-
nem necessitate fieri, sed non omnem quae necessitate fit facere aequivocationem, nisi
per impositionem stabile ei illud nomen imponatur.’
The De Generibus et Speciebus and the Theory of Collectio 187
38 We also find similar passages in Abelard; see for instance the text just quoted of Dialectica
112–114
39 De Gen. 104 (164), 22–24: ‘Nota quod nomina illa tantum dicuntur substantiva quae impo-
nuntur ad nominandum aliquem propter ejus materiam, ut homo et caetera universalia
substantiva.’
188 Chapter 7
example of man with its individual instances (where as1 ,….., asm are the fea-
tures identifying Socrates):
We can hardly think that being a rational animal philosopher is the essence of
Socrates, but it is not particularly troubling to call such constituent material
part of Socrates. From this perspective, the essence would be a specific mate-
rial constituent. It does not, therefore, seem implausible that Joscelin spoke
mainly of essences simply because he was most interested in solving the prob-
lem of the species and genera of substances.
In the period with which we are dealing, numerous discussions can be found
on the meaning of the verb ‘to be’, both grammatical and dialectic. Abelard, for
example, reports two different trends: some believe that the copula indicates
a relation of identity, while others think it expresses the inherence between
the meaning of a predicate and the individuals denoted by the subject. The
‘winning’ theory, which is accepted in late scholasticism, is that of identity.
According to this approach, a categorical sentence is true if and only if any
individual denoted by the subject term is identical to some individual denoted
by the predicate expression. In the opinion of the supporters of the theory
of inherence, on the other hand, a categorical statement is true if and only if
The De Generibus et Speciebus and the Theory of Collectio 189
when and only when the individual essence of Socrates is contained in ‘being
man’ (at some point in time), but it is not clear whether we can say the same
for rationality.
Joscelin should say something more on this point, if he wishes to suggest a
correspondence between the analysis of a categorical statement with a com-
mon noun as a predicate and one with an adjective in the same position. The
following could be a ‘transparent’ formula with reference to the truth condi-
tions for sentences with an adjective in predicate position: ‘∃f Π(f, being ratio-
nal) ∧ Por(f) = Socrates’, i.e., there exists a rationality-component possessed by
Socrates. Π expresses a relation between a collective form and its qualitative
components.
After presenting his theory, Joscelin tries to show how this agrees with reason
and authority. We label as r1–r10 the remarks which include both clarifications
and answers to possible objections.
r1. It is said that the species /man/ is the matter of individual men, but
the species, understood as a collection, cannot be the matter of individuals.
Joscelin’ reply is rather ingenious: the species is a collection of individual es-
sences. When it is said that the species /man/ takes on the specific property of
Socrates-ness, thus becoming the matter of Socrates, what is meant is that the
whole takes on the property received by the part:
43 De Gen. 92 (160), 18–25: ‘Illud tantum humanitatis informatur Socratitate quod in Socrate
est, ipsum autem species non est […] Attende: materia est omnis species sui individui et
ejus formam suscipit, non ita scilicet quod singulae essentiae illius speciei informentur
illa forma, sed una tantum […] quia similis est compositionis prorsus cum omnibus aliis
ejusdem naturae essentiis, quod ipsa suscipit compactum ex ipsa et caeteris suscipere
autores voluerunt.’
The De Generibus et Speciebus and the Theory of Collectio 191
In order to clarify how one totality can accommodate different forms, Joscelin
gives the example of a mass of iron that is shaped in such a way as to yield a
knife or a nail. Besides clarifying the point, the example provides a further clue
as to how Joscelin understands collections.
r2. The species is essentially what is predicated (in quid) of several individ-
ual things. According to Joscelin, to say that the species is predicated is to say
that it is ‘inherent in’ a certain individual. The problem is not only terminologi-
cal, it is in fact a matter of understanding the relationship between the species
(genus) and its individuals. The idea is that the individual has a relation-
ship with an essence which composes a multitude: Socrates’ human essence
touches (tangit), so to speak, the corresponding essence of the species /man/.
Joscelin adds other examples in an attempt to make the point more plausible:
I say that I touch the wall not because the individual parts of me are in
contact with the wall, but only because the tip of my finger may perhaps
be touching it; similarly an army is in contact with a bastion or with a
place [ …]. In the same way as the species, although the identity of some
essence of that collection with the whole is greater than that of a person
compared to the army.44
44 De Gen. 96 (162), 7–12: ‘Dicor tangere parietem non quod singulae partes mei parieti haere-
ant, sed forsitan sola sommitas digiti dicor tangere. Eodem quoque modo exercitus aliqu-
is dicitur haerere muro vel alicui loco […] Similiter de specie, quamvis major sit identitas
alicujus essentiae illius collectionis ad totum quam alicujus personae ad esercitum.’
45 Abelard criticizes a theory of collectio that is clearly different from this one, where univer-
sals are collections of individuals.
192 Chapter 7
not sets of individuals (in the standard or mereological sense of the term), nor
collections of properties.
r3. The species is predicated in quid (that is, in essence) of several numeri-
cally different things. In some sense, Joscelin says, to be predicated in essence
means that the predicated thing is identical with that of which it is predicated.
If this applies to species/genera, singular sentences would express a relation
of identity between particulars and universals. As this is not possible, we must
consider essential predication in the sense of the Categories: what is said-of
and is not-in (i.e. is not an accident) is a universal essential predicate. The focal
point of this observation concerns the meaning of the predicative link. As we
have seen, Joscelin has no sympathy for the identity theory; he believes that in
a categorical true sentence like ‘homo est animal’, the verb ‘to be’ expresses a
relation between specific human essences and a collection of generic essences.
r4. Species change by adding or removing essences. In Joscelin’ view, es-
sences are in some way associated with individuals, so if an individual is born
or dies, an essence is born or dies, and thus the species/genus of which that es-
sence is a part changes. However, species/genera are also meanings of common
nouns. It follows, then, that if an individual is born or dies, the meaning of the
noun for its species/genera changes. This difficulty is formulated by Abelard,
although his criticism seems to be more directed against a theory concerning
collections of individuals rather than the position we are discussing here.
Revealing remarkable logical acumen, Joscelin establishes a connection be-
tween meanings and points in time: according to his theory, a sentence like
‘Socrates est homo’ is true at a certain time if Socrates’ human essence is one of
the essences that compose the human species /man/ at that moment of time,
and is false if the same essence is not a component of the species at that mo-
ment. As to the difficulty raised by Abelard, Joscelin’ reply is as follows:
Today, when ‘Caesar beat Pompey’ is said, one has a concept of the
same thing as when Caesar lived. Today, however, Caesar is not Caesar.
Similarly ‘man’ names something made up of being-man, that is of hu-
manity, but whether it is formed by humanity consisting of ten or more
essences does not depend on the meaning of the expression. The state-
ment ‘Socrates is a man’ is just as true as ‘Socrates is formed by humanity’,
for any number of essences of which humanity is made up.46
46 De Gen. 115 (170), 4 – 116 (170), 21: ‘Nisi singulis momentis significatio hujus vocis homo
mutetur, non potest vere dici bis: Socrates est homo. Nam cum iterum dixeris : Socrates
est homo, si dicas esse de humanitate quam prius dixeris, falsum est, nam illa jam non
est. Attende : verum est quidem quod illa humanitas quae ante mille annos fuit vel quae
haeri, non est illa quae hodie est, sed tamen est eadem cum illa, id est creationis non dis-
similis. Non enim quicquid idem est cum alio, idem est illud ; homo enim et asinus idem
sunt in genere, nec tamen hoc est illud. Socrates quoque ex pluribus atomis constat vir
quam puer, et tamen idem est. Vocis quoque significatio non mutatur quamvis hoc non
sit illud, ut patet in hac voce Caesar quae idem significat mortuo Caesare, quamvis non
sit verum dicere : Caesar est Caesar ; cum enim dicitur hodie : Caesar vicit Pompeium, de
eadem re habetur intellectus de qua vivente Caesare ; hodie tamen Caesar non est Caesar.
Similiter homo nominat aliquid materiatum ab homine, scilicet humanitate, sed non ex
vocis significatione est utrum ex humanitate constante ex decem sive ex amplioribus.
Tamdiu ergo verum est dicere : Socrates est homo, quamdiu est materiatum ab humani-
tate, ex quantislibet essentiis humanitatis constante.’
47 I have argued elsewhere that some version of the Montague Grammar can be useful
in interpreting ‘conservative’ changes of the meaning of general terms: Cf. R. Pinzani,
The Logical Grammar cit., 141–42.
194 Chapter 7
and a second relation connecting these same occurrences between them. The
second relation can be considered as a kind of causal connection, in the sense
that there exists a generative-corruptive process connecting Socrates’ various
occurrences.
Even in the case of collections of essences, the plan of creation presents
itself as a network of relationships. Joscelin appeals to the relation of similar-
ity to define a certain collection of objects. In the case of genera and species
Joscelin says that similar collective totalities are to be considered instances of
the same universal. We have no causal relationship in the sense of a generative-
corruptive process. A different relation connects, on the one hand, different
moments and situations to singular collections of essences, and, on the other,
the singular collections of essences to individual owners of such essences.
Joscelin’ theory, therefore, does not incur the same difficulties raised by
Abelard about a version of the collection theory. According to what we read
in the Logica Ingredientibus, those who identify universals with collections is
required to explain why universals remain unchanged despite the variation in
the extension of terms. As we have seen, Joscelin’ response fulfills the request.
What may not be immediately obvious to modern interpreters is that in criti-
cizing the supporters of the collectio, Abelard also criticizes any theory which
associates extensional entities to common nouns. The theory of translatio al-
lows Abelard to keep the idea of an invariant meaning and at the same time
to give an account of the different evaluations of a sentence. Joscelin actually
follows the same path; despite introducing the theme of translatio mainly to
distinguish between direct and indirect meaning.
r5. The species is predicated of many things. One wonders how this com-
munis opinio agrees with the fact that there may be more collections of human
essences. Joscelin notes that ‘to be predicated’ means ‘to be signified by the
expression which plays the role of predicate’. Here the criticism is similar to the
first objection that Abelard makes to the proponents of the collectio theory. If
universals are collections, any grouping of individuals constitutes a universal.
The same criticism can be reformulated by considering collections of essences
instead of individuals. Against the objection that there are too many collective
universals, Joscelin observes, on the one hand, that there is only one humanity,
on the other, that there are as many universals as the plan of creation allows.
The first answer refers to the solution just discussed. The second evokes cases
such as ‘men in the room’, which do not express genuine universals, but rather
(if the analysis of signification we have provided is correct) signify objects of
the same sort of universals. The difficulty underlying the second observation
is more general and includes any reconstruction of the Aristotelian relation
of predication. We can talk about the plan of creation in metaphorical terms,
The De Generibus et Speciebus and the Theory of Collectio 195
or in logical terms we can say that there are modal restrictions allowing us to
select the ‘right’ species and genera.
r6. Do species with only one individual exist? According to Joscelin’ theory,
there exist species with one individual only which does not coincide with the
species. It would certainly be problematic to distinguish between a totality
formed by a single part and the same part. Given that the species is a multiplic-
ity of essences, however, the distinction between an essential component and
the individual constituted by such a component plus a differential element is
clear.
r7. The essence of man present in me must be a substance, but it does not
appear to be a first substance (since only the individual is a first substance) nor
a second substance (as it is something particular and not universal). Joscelin
replies that it is a substance sui generis, which does not fall in the distinction
drawn by Aristotle in the Categories.
r8. According to Boethius, species do not take a part of the genus but at
the same time have all the genus. This would seem to go against the argu-
ment that /man/ is constituted of part of animal essences plus rationality (or
rather, rationalities); consequently, the whole animal is not entirely present
in the individual species. Joscelin suggests that Boethius’ words should be
put into the particular context of his philosophy, then proposes the following
interpretation:
When Boethius denies that species take a part of the genus, he does not
speak of the essences that compose those multitudes, but of the final
parts, i.e. the genus /animal/ consists of /body/ as matter and /being sen-
tient/ as form. Therefore, when it changes, by parts of its quantity, into
species, it does not grasp the matter of the species without form, and
different form without matter, but in each species there is the matter and
form of the genus.48
Joscelin seems to be saying that ‘having all of the genus’ means having at the
same time a material and a formal component. In other words, when Boethius
says that all genus is present in this species, he means that formal and material
parts are all together in the species.
48 De Gen. 138 (178), 16–21: ‘Cum negat Boethius species partes generis carpere, non de es-
sentiis illam moltitudinem conjungentibus agebat, se de partibus diffinitivis. Verbi gratia,
animal genus ex corpore constat materia, ex sensibilitate forma. Cum ergo per partes suae
quantitatis transit in species, non arripit una de speciebus materiam et non formam, et
alia formam et non materiam, sed in singulis speciebus materia et forma generis est.’
196 Chapter 7
Plato believed that the substance itself was made by the admissibility of
contraries as form and the mere essence as matter […]. In Socrates there
is a part of mere essence […] the admissibility of contraries conforms
such mere essence and from these things a certain essence of the sub-
stance results […] (at the end of the specification process) the Socrates-
ness shapes that essence of man and realizes Socrates. At the same time,
heat shapes other atoms of its essence of man, and forms of fire realize
the fire, other forms of water shape water, and air forms the air or earth,
and so the individual particles are fire, water or earth.50
50 De Gen. 149 (186), 5–6; 156 (188), 12–33: ‘Ipsam (substantiam) invenit (Plato) constare
ex susceptibilitate contrariorum forma, materia autem mera essentia […] Est igitur in
Socrate quaedam pars merae essentiae […] susceptibilitas contrariorum; eam informant,
et ex his constituitur quaedam essentia substantiae. […] Socratitas totam illam essentiam
humanitatis informat et Socratem facit. Tam statim vero alias atomos illius essentiae hu-
manitatis afficit calor et formae ignis et ignem faciunt, alias formae aeris et aerea faci-
unt, alias terrae et terrma faciunt, et sic singulae particulae vel ignis sunt vel aqua vel aer
vel terra’.
chapter 8
In dealing with Abelard’s dialectic one needs to account for several distinct yet
intertwined issues. These issues have to do with grammar, the theory of mean-
ing, logic intended as theory of proof, and metaphysics. The net is cast where
medieval philosophers fish, namely, in the waters of authoritative texts. During
the period when Abelard writes his works, these texts – essentially those of
logic vetus – could easily have fit inside an armarium (and it is possible that
they actually did, in the room that the Palatine Philosopher reserved for his
meditations.
I will focus on the writings devoted to the Porphyrian problem. My recon-
struction of Abelard’s position on the problem of universals is essentially
based on the commentaries on Porphyry in the Logica Ingredientibus and the
Logica nostrorum petitioni sociorum.1 In these works Abelard does not abide
by the genre rules of the literal glosses and provides lengthy and independent
doctrinal developments. The reader is invited to reflect upon the fact that in
addition to the problems raised by Porphyry, “there are other no less demand-
ing issues, such as the common cause of imposition and the conceptual mean-
ing.” This observation seems on the one hand to indicate the urgency of certain
important themes, and on the other a somewhat unconventional style.2
1 On the texts and their dating , see: C.J. Mews, ‘On Dating the Works of Peter Abelard,’ Archives
de l’histoire doctrinale et littéraire de moyen age, 52 (1985), 73–134; De Rijk, ‘Peter Abaelard’s
Semantics and His Doctrine of Being,’ Vivarium 24 (1986), 85–127; M.T. Clanchy, Abelard: a
medieval life, Blackwell, Oxford 1997 (Cf. in particular pp. 103–104); J.E. Brower – K. Guilfoy,
‘Introduction,’ in J.E. Brower – K. Guilfoy (edd), The Cambridge Companion to Abelard,
Cambridge UP 2004, 1–12; J. Marenbon, ‘Life, Milieu, and Intellectual Contexts,’ ibid., 13–44;
K. Jacobi, ‘Philosophy of Language,’ ibid. (in particular pp. 129–130); P. King, ‘Peter Abelard,’
in Stanford Encyclopedia of Philosophy (par. 1.2). Recently scholars have raised doubts as to
the attribution of the Nostrorum to Abelard; for instance Marenbon, in maintaining that the
material in this work is largely Abelardian, believes that ‘it would be justifiedly skeptical’ to
have concerns about the authenticy of the second part of the work (Abelard in four dimen-
sions. A twelfth-century philosopher in his context and ours, University of Notre Dame Press,
Notre Dame 2013, p 38).
2 Cf. Log. Ing., 8, 11–14. This point is duly noted by P. King, P. King, ‘Abelard’s answers to
Porphyry,’ Documenti e studi sulla tradizione filosofica medievale, 18 (2009), 249–270.
This argument is, on the one hand, reminiscent of the maxim contained in the
second Boethian commentary on Porphyry (what exists is one, what is com-
mon does not exist); on the other, of Boethius’ observations on the part-whole
3 The first two chapters of our The Logical Grammar cit. are devoted to this contamination and
the topic of categorization.
4 Log. Ing., 10, 9–15: ‘Nulla enim res nec ulla collectio rerum de pluribus singillatim praedicari
videtur, quod proprietas universalis exigit. Nam etsi hic popolus vel haec domus vel Socrates
de omnibus simul partibus suis dicatur, nemo tamen omnino ea universalia dicit, cum ad
singularia praedicatio eorum non veniat. Una autem res multo minus quam collectio de plu-
ribus praedicatur.’
200 Chapter 8
– x is part of Y iff x is a Y (by substitution: the roof is part of the house iff the
roof is a house).
TR1
a. The universal is a singular thing
b. It constitutes the essence of several individuals.
c. The essence is different for the forms of ‘inferior’ things ( formae
inferiorum).
d. If accidental differences are subtracted, things are indistinguishable.
The second thesis attributed to ‘realists’ is more general, and may be shared by
philosophers of different orientations. The sub-theses common to the differ-
ent positions are the following:
TR2
a. Individual things are distinguished by their forms and essence
b. There exists nothing which is in several individual things at the same
time
c. If the accidental forms are subtracted, things remain individual
5 See in particular De Divisione 8; 38 (J. Magee edition, Brill, Leiden – Boston – Köln 1998).
The Position of Abelard on Porphyry 201
TR2.2 in turn is specified in slightly different ways which may or may not re-
flect diversified realist positions:
1. If the same thing is present in several different things, then the animal
formed by rationality is the animal formed by irrationality
2. Therefore, in the same thing there are contrary determinations.
The rational animal is different from the irrational one, but the component
‘animal’ nested in the rational animal is the same as the one nested in the ir-
rational one. Accordingly, the same substantial component ends up being sub-
jected to contrary determinations. Abelard formulates the classic anti-realist
argument showing how contraries cannot be simultaneously present in the
genus.
As we have seen in previous chapters, the question of the presence of
contrary properties in the genus gives rise to several difficulties, more or less
evident depending on the theoritical framework. Realist philosophers were
unimpressed by this first criticism, and maintained that the presence of op-
posites in the genus or in the species was not as problematic as in individuals.
Abelard moves on to show that Socrates and Brunellus are Socrates, and there-
fore rationality and irrationality are in Socrates. The argument runs as follows:
reach the inner core of the substance (a case Abelard considers later); at that
layer all things will be equal in that they contain the same material essence.
It is important to understand the argumentative strategy. The point cur-
rently criticized is the presence of the universal in the particular. In order to be
acceptable for its opponents the argument concluding with the thesis in ques-
tion must start from shared premises. Premise (b) needs to be demonstrated,
and is not directly shareable among realists. Premise (a) remains, which might
be acceptable if understood as a reformulation of TR1.
Abelard tries to see how (b) is true by exclusion, since the following theses
are valid:
1b. Brunellus’ forms are not Brunellus (since the accidents would be precisely
the substance)
2b. matter and forms together are not Brunellus.
Thesis 2b is very difficult to understand (and accept) if the verb ‘to be’ is taken
as a sign of identity, because it contrasts with a cornerstone of Aristotelian
metaphysics, namely the essential composition of first substances. As such, it
is important to understand how it is that a realist might accept such an argu-
ment, even before recognizing its paradoxical consequences (which are read-
ily apparent). If we continue the metaphor of the onion, the example of the
body would mean that if we remove from Socrates certain layers reaching the
essence of the body, at this point Socrates is that body, and no longer the com-
pound of the body and forms (since we have removed a certain number of
forms).
In the books on Topic in the Dialectica, there are some points that may help
us to understand thesis 2.b. Abelard believes ‘animal est homo’ is acceptable
on the basis of an interpretation that makes explicit the clause determining
the subject, that is to say, ‘quod in homine est, idest rationalitate et mortalitate
informato.’Therefore, the sentence should be rewritten as ‘animal <quod in ho-
mine est> est <idem quod> homo.’9 In the Logica Nostrorum Abelard illustrates
(among others) a theory of the universal intended as ‘sermo’, proposing a list of
opposing theses whose truth appears to vary depending on the position of the
essences in Socrates, as we are not obliged to maintain that there are two substances in act in
a chair (e.g. the armrest and the seat).
9 Cf. Dial. 415, 28–32: ‘neque enim animal quod in homine est, idest quod rationalitate et mor-
talitate est informatum, res est alia ab homine, nec ligna aut lapides quae in constitutione
domus aliud putanda sunt a domo. Quae enim essentiae partes sunt, omnes simul acceptae
idem prorsus sunt quod totum’ (ibid. 28–31).
204 Chapter 8
terms.10 The passage is rather complicated, and as we shall see, it has aroused
some interest in historiography. From a certain point of view Abelard consid-
ers the statement ‘haec materia Socratis est Socrates’ acceptable, but prima
facie ‘est’ does not act as a sign of identity, since ‘Socrates est (idem quod) haec
materia Socratis’ is unacceptable. This interpretation may play some role in the
case we are discussing.
If there is a sense in which we speak of the essence of Socrates apart from
its forms, and this essence is a universal, then we can talk about a variety of
abstract essences coinciding with individuals without their forms. However,
the consequences would be unacceptable for Abelard. Whatever the argument
may be, it seems that not all realists are willing to grant 2.b. The point of con-
tention seems to be whether we can speak of individuals regardless of their
forms. Abelard’s counterargument is quite clear:
It is not even possible to maintain what [realists] assume, i.e. that in-
dividuals are formed by their accidents. In fact, if individuals owe their
being to accidents, then naturally the accidents precede them, just like
differences precede species. Indeed, they say that as /man/ is different
[from another species] through the substantial action [ex formatione] of
difference, so Socrates is distinguished [from other individuals] because
of accidents. Therefore, Socrates cannot be without accidents nor man
without differences. For this reason Socrates is not their foundation, just
like man is not the foundation of differences. If, on the other hand, ac-
cidents are not in an individual substance as in their foundation, neither
are they in the universals.11
to assume that Socrates can do without human form or man can be without
rationality is more challenging.
I would like to emphasize the fact that this first part of the commentary
on Porphyry does not conclude that there exist no second substances such as
/man/ or /animal/, but rather that, if these substances exist, they are not uni-
versal. Abelard continues on to criticize the theories that identify the universal
with collections and individuals. Still, it is important to remember that the goal
of the criticism is not to question the fact that there are collections, but that
such entities are the universals mentioned by Porphyry. It is no coincidence
that the first criticism is introduced as follows:
We will now refute the collection thesis, according to which the entire
collection of men, which is considered a single species, can be predicated
of several things at the same time, being a universal.12
The refutation is divided into two parts. In part one, the collection is consid-
ered as a physical totality; according to this view, to say that a certain totality
is predicated of something is equivalent to saying that it is physically present
in its components. In part two the collection is seen as a set or mereological
totality (the two points are not clearly distinguished) and the predication rela-
tion is one-many; it is not required that the one is ‘in’ the many. The criticism
focuses on the multiplicity of objects that can be called ‘collections’ and on the
fact that the elements (parts) precede the whole.
In (1) Abelard shows that a whole, understood in the physical sense, does
not meet the definition of universal, since the latter must be present at the
same time in all the individual things of which it is predicated. In addition,
any physical body might be considered a whole with respect to its sub-parts,
for example, Socrates himself would be a universal in relation to his hands,
feet, etc. The criticisms in (2) are more general and do not depend strictly on
the interpretation of predication as physical presence of the whole in its parts:
2.1 Any collection of things could be a universal. The species /man/ would
no longer be the last species, but rather would have different species
under it corresponding to different possible groupings of men. This diffi-
culty arises if we assume that the relation of predication is intended in
12 Log. Ing., 14, 32–34: ‘Nunc autem prius infirmemus sententiam quae prior posita est de
collectione et quomodo tota simul hominum collectio quae una dicitur species de pluri-
bus praedicari habeat, ut uneversalis sit.’
206 Chapter 8
accordance with the essential definitions and the terms of these defini-
tions are given once and for all.
2.2 ‘Any collection of bodies and spirits would be a universal substance, just
like the entire collection of substances would be a very general totality,
and consequently there would be several very general collections of
substances.’13 This probably means that, considering some (but not all)
instances of substance, the corresponding nodes of a Porphyry’s tree
would be labelled substance. However, according to the traditional way
of conceiving the predication tree, different nodes must have different
labels, and there is only one apical node labelled substance.
2.3 It cannot be objected that there are more general and less general collec-
tions of substances, since they should be called differently; a proposal
that goes against the implicit assumption on the limited number of labels
and other restrictions on the relation of essential predication.
2.4 By nature, the universal comes before its individuals, while the collection
as a whole is posterior to its parts.
2.5 The genus (plus differential properties) is equal to the species, but the
part (plus some properties) is not equal to the whole.
The opinion of those who say that individuals are universal because they
are like others remains to be refuted. These philosophers say that the
same individuals are predicated of many things, not because they are es-
sentially several things, but because several things are like them.14
1. The individual is predicated of one thing only, so it is not like other things.
However, it has been assumed that the individual is like other things; is the
common doctrine wrong?
2. How can a distinction between universal and singular be maintained, if
/man/ is similar to many things just as Socrates does? If /man/ behaves like
Socrates, then both either are similar or are not.
3. If /man/, which is in Socrates, is the same thing as Socrates, there is no dif-
ference between the two. I would say that this criticism is the position of
common sense. Only at the cost of a complicated exercise can one make
sense of the theory of likeness-identity.
4. How can the sentence ‘Socrates and Plato are alike in being man’ be under-
stood, since Socrates (and therefore man as coincident with Socrates) dif-
fers completely from Plato? A variant of the theory under discussion is
considered, according to which ‘in homine conuenire negatiue accipiunt, ac
si diceret<ur>: non differt Socrates a Platone in homine.’ Abelard’s comment
is rather sarcastic: ‘Sed et sic quoque potest dici quod nec differt ab eo in
lapide, cum neuter sit lapis.’15
5. If instead one wants to say ‘Sunt homo, quia in homine non differunt,’ in the
sense that Socrates does not differ from Plato in his being identical to
/man/,16 Socrates does not differ from Plato in himself either (which is
absurd); if he is different from Plato, he is also different from /man/
(against the thesis that has been formulated).
Having explained the reasons why things, either considered in their sin-
gularity or gathered together, cannot be said to be universal (since uni-
versals are predicated of several things), all that remains is to attribute
universality to linguistic expressions only. In fact, just as grammarians
call some nouns appellative and some proper, similarly some simple
Abelard’s says that ‘to be predicated’ means ‘to be truly connectable to some-
thing by means of the substantive verb (‘to be’) conjugated in the present
tense’. Even intransitive verbs, such as ‘run’ and ‘walk’ can be predicated of
many things, or rather, being equivalent to a verbal expression (in the form ‘is
17 Log. Ing., 16, 19–29: ‘Nunc autem, ostensis rationibus quibus neque res singillatim neque
collectim acceptae universales dici possunt in eo quod de pluribus praedicantur, restat
ut huiusmodi universalitatem solis vocibus adscribamus. Sicut igitur nominum quaedam
appellativa a grammaticis, quaedam propria dicuntur, ita a dialecticis simplicium sermo-
num quidam universales, quidam particulares, scilicet singulares appellantur. Est autem
universale vocabulum quod de pluribus singillatim habile est ex inventione sua praedi-
cari, ut hoc nomen ‘homo’, quod particularibus nominibus hominum coniungibile est se-
cundum subiectarum rerum naturam quibus est impositum. Singulare vero est quod de
uno solo praedicabile est.’
The Position of Abelard on Porphyry 209
TDAbelard.a ‘x is (an) X’ is true iff what the sentence says is a fact (illud
quod dicit propositio est in re)
TDAbelard.b ‘x is (an) X’ is true iff the meaning of ‘x’- the meaning of
(status) ‘X’ is a fact
where ‘-’ stands for some operation or application (of one meaning to the
other) yielding the meaning of the sentence. From TDAbelard.b and P1Abelard.a
we get:
18 Log. Ing., 17, 12–21: ‘Nota autem aliam esse coniunctionem constructionis quam attendunt
grammatici, alia<m> praedicationis quam considerant dialectici. […] Praedicationis uero
coniunctio quam hic accipimus ad rerum naturam pertinet et ad ueritatem status earum
demonstrandam.’
210 Chapter 8
P1Abelard.c ‘X’ ℘ ‘x’ if and only if the meaning of ‘x’- the meaning
(status) of ‘X’ is a fact.
P1Abelard.d the meaning of ‘x’- the meaning (status) of ‘X’ is a fact if and
only if x is an X.
the things from which we have started. This had obviously been understood
right away, and Abelard realizes he owes some explanation:
The criticism of the realist thesis shows that there are no universal things.
The point is rather subtle: universal expressions are grammatically different
from particular expressions; so, the meaning of universal expressions – if any –
should somewhat differ from that of proper nouns:
It does not seem that there are things upon which a universal expression
could be imposed (imponi),20 since all things are particular, and as we
have seen, they do not share anything real, according to which universal
nouns can be imposed. It is certain that universal expressions are not im-
posed on things according to their peculiar difference (in that case they
would not be common but singular), and at the same time they cannot
name particular things as if the latter share some entity (indeed there is
nothing that they share), so it seems that universal expressions have no
real meaning.21
Things, therefore, are all individual and share nothing. This is reminiscent of
the criticisms of the realist theses just considered. The text continues:
19 Log. Ing., 18, 5–9: ‘praecipue uniuersalium uocum proprietatem diligenter perquiramus.
De quibus uniuersalibus positae fuerant quaestiones, quia maxime de earum significa-
tione dubitatur, cum neque rem subiectam aliquam uideantur habere nec de aliquo intel-
lectum sanum constituere.’
20 The term ‘impositio’ plays an important role in the development of medieval semantic
theory. For a first look, see my ‘The Origin of Language in Medieval Logic Commentaries,’
in F. Amerini – R. Messori (eds.), Sulle origini del linguaggio. Immaginazione, espressione,
simbolo, ETS, Pisa 2012, 79–91.
21 Log. Ing., 18, 9–17: ‘Rebus autem nullis uidebantur imponi uniuersalia nomina, cum scili-
cet omnes res discrete in se subsisterent nec in re aliqua, ut ostensum est, conuenirent,
secundum cuius rei conuenientiam uniuersalia nomina possint imponi. Cum itaque cer-
tum sit uniuersalia non imponi rebus secundum suae discretionis differentiam, quippe
iam non essent communia sed singularia, nec iterum eas possint ut conuenientes in ali-
qua re nominare; quippe nulla res est in qua conueniant, nullam de rebus significationem
contrahere uidentur uniuersalia.’
212 Chapter 8
In fact, ‘man’ is imposed on things for the same reason, namely the fact
that they are mortal rational animal […] in the common noun ‘man’ one
does not understand either Socrates or anyone else or the entire col-
lection of men, nor, as some believe, with that noun does one refer to
Socrates as a man.22
A universal term like ‘man’ is imposed on individual men due to the fact that
they are rational mortal animals. This is the core of Abelard’s semantic theory.
Two key phrases are present in the text: ‘[common] cause of imposition’ and
‘being rational mortal animal’. The former is used as a synonym of ‘status’
(term designating the meaning of a general noun), the latter is the name of
the meaning of ‘man’. The semantic theory outlined is dualistic: expressions
signify particular things and, besides, a sui generis property. Towards the end
of the passage Abelard points out that when we say ‘man’ one does not under-
stand (and thus, does not mean) Socrates, or the collection of men, or Socrates
himself as a man, i.e. none of the candidates for the role of universal criticized
above. What one understands is a something according to which common
nouns denote things.
So far Abelard has simply claimed that universal things do not exist and
therefore the status is not a universal thing. In the caveat that follows he goes
further, arguing that status is not a thing:23
Being man is not a man nor some other thing, if we look carefully, just like
not being in a subject, or not allowing contraries, or not allowing more
22 Log. Ing., 18, 27–30: ‘Nam quoniam ‘homo’ singulis impositum est ex eadem causa, quia
scilicet sunt animal rationale mortale […] in nomine uero communi quod homo est
nec ipse Socrates nec alius nec tota hominum collectio rationabiliter ex ui uocis intel-
ligitur nec etiam ipse Socrates in quantum homo est per hoc nomen, ut quidam uolunt,
certificatur.’
23 The ambiguity of the concept of ‘thing’ should be noted. Jolivet pointed out that the term
‘res’ is used by the philosopher with rather different meanings (J. Jolivet, Aspects de la
pensée médiévale: Abélard. Doctrines du language, Vrin, Paris 1987, 128–132). The expres-
sions ‘res speciei and ‘res generis’ are equally equivocal (Cf. Log. Ing. 72, 4–8; 78, 16–17; 85,
5–7; ibid. 18–20; 99, 7–15; etc.), but it seems appropriate to consider the term ‘res’ in the
singular, as for example in ‘res dicitur genus quae est essentia diversarum specierum’ (Log.
Ing. 125, 16). In the realist interpretation, ‘translated’ into Abelard’s language, these res
are called ‘common (signified) natures’ or ‘status (signified)’. In a similar sense Priscian,
quoted by Abelard (Log. Ing. 23, 26–27), speaks of the species and genus as proper nouns.
The Position of Abelard on Porphyry 213
or less, are not things, yet however they are that in which, according to
Aristotle, all things are alike (convenire).24
In order to explain why being a man is not a thing, Abelard uses the example of
negative or privative terms. The example has an important history that dates
back to the Carolingian age at least. The standard interpretation in Abelard’s
era is the one provided by St. Anselm, according to whom a privative term
has a positive meaning and a way of signifying which consists in removing the
positive meaning. These examples, however, do not seem to have been well
chosen. If we admit that to be in a subject or to not allow contraries is to not
be something real, we may continue to believe (assuming that the standard
theory is correct) that being in a subject and allowing contraries is being some-
thing. Since there is no negation in ‘being man’, however, what happens in this
case is unclear. If we remove ‘being’, we are left with the term ‘man’, which, ac-
cording to Abelard, means ‘being man’.
If the sense in which being man is not to be considered a thing is somewhat
unclear, what is clear is Abelard’s concern with finding a few examples that
help us to imagine a new kind of entity. The concept of likeness in this context
plays a key role. The fact that individuals x and y share the property φ does not
imply that x, y ‘contain’ or are the same as φ. The background physicalist theory
does not apply, since it is possible to be alike in not having a property, and that
which things do not have cannot be in them in any way.
The doctrine of the common cause of imposition is not entirely stable.
Although they do not exist, statuses do many things:
1. they are the cause due to which universal expressions are imposed on
things,
2. they are something shared by several things,
3. they can be objects of imposition despite their not existing,
4. they have causal power, just like being sick is cause of death 25
24 Log. Ing., 19, 25–29: ‘Esse autem hominem non est homo nec res aliqua, si diligentius con-
sideremus, sicut nec non esse in subiecto res est aliqua nec non suscipere contrarietatem
uel non suscipere magis et minus, secundum quae tamen Aristoteles omnes substantias
conuenire dicit.’
25 Points 1–2 have already been considered. Another interesting text is Log. Ing., 20, 7–14 :
‘Statum autem hominis ipsum esse hominem, quod non est res, uocamus, quod etiam
diximus communem causam impositionis nominis ad singulos […] Statum quoque homi-
nis res ipsas in natura hominis statutas possumus appellare, quarum communem simili-
tudinem ille concepit qui uocabulum imposuit.’ As for point 4., we have seen some texts,
others will be seen and discussed in the next paragraph.
214 Chapter 8
Point (2) is particularly interesting. Abelard does not want one to think that
because statuses do not exist, things share nothing; for this reason he points
out that it is no different to say that many men are alike in being men and that
many men are men. But in doing so he makes similarity a fictitious concept, a
figure of speech that can be abandoned, as the equivalence ‘x, y are men iff x,
y are alike in being men’ holds. The historiographical assessment of Abelard’s
position may vary depending upon the emphasis one places upon the different
interpretations of the Aristotelian principle of predication.
The passages in the Categories and Isagoge that appear to involve ontologi-
cal questions are systematically interpreted in the Ingredientibus as analyses
of the meaning of terms. This is the third exegetical option Abelard claims to
have discovered, as an alternative to both the extreme nominal interpretation
and the positions of the realists. The same point of view is present in the Logic
Nostrorum; the reader is immediately informed of the fact that when speaking
of ‘the five’ (genus, species, difference, property and accident), authorities can
refer to either nouns or meanings. Abelard begins with the Aristotelian defini-
tion of universal and particular:
26 Log. Nostr., 512, 10–22: ‘Ab uniuersalibus inchoandus est, definiendo scilicet quid pro-
prie uniuersale, quid singulare (id est particulare siue indiuiduum), dicendum sit.
Uniuersale aliter Aristoteles describere dicitur in Categoriis ‘quod de subiecto dicitur’ uel
in Perihermeneias ‘quod natum praedicari de pluribus’. Similiter et indiuiduum cum in
Categoriis, ut ‘quod non dicitur de subiecto’, in Perihermeneias uero ‘quod non est aptum
natum praedicari de pluribus, immo de uno solo’. Cum diuersi has diuersas definitiones
The Position of Abelard on Porphyry 215
Here, then, we see three different points of view from which universals can
be considered. Abelard observes that, in certain texts, the authorities explic-
itly refer to things that correspond to grammatical predicates; he then notes a
number of possibilities:27
Some things are higher or General forms are first in Genera and species mean
more universal (Boethius) God’s mind (Priscian) quality (Aristotle)
/man/, the thing itself, Genera and species The genus expresses
consists of genus and are concepts in the the common substance
difference (Boethius) divine intellect (a thesis (Boethius)
attributed to Plato)
Some things are said of or The genus is part of the
are in others (Aristotle) definition (a thesis shared
by all philosophers)
The most convincing examples are the ones demonstrating a real consider-
ation of genera and species. Curiously, the conceptualist position is document-
ed only by reference to ideas present in God’s mind. Abelard believes that the
passages in the right-hand column represent a linguistic interpretation: in the
first two quotes (on the right) the focus is on the relation of meaning, and in
the third a reference is made to the definition, and not to the proposition ex-
pressed by it. ‘Sermo’ is a technical term which indicates a meaningful expres-
sion (i.e. not merely a speech sound or graphic sign).
The first realist argument considered is TR1. The reader may recall that
the universal is regarded as a thing present in particular objects. In the
Ingredientibus there was a line of reasoning that led to the identification of
Socrates and Brunellus; here, Abelard says that the presence of a specific es-
sence in several individuals involves their identity. This passage is not easy to
interpret:
ad diuersa applicent, alii uidelicet ad res, alii uero ad intellectus, alii ad sermones, tamen
unusquisque tuetur se auctoritate iudice.’
27 Log. Nostr., 512–515 passim.
216 Chapter 8
material essence x is in a
material essence x is in b
if the same material essence is in a and b then a = b
a=b
Abelard claims that if we removed the forms, individuals would not differ. This
sounds vaguely tautological: if from two things we take away what differenti-
ates them, then clearly the two become one. An effective criticism must state
something more, possibly asserting the antecedent of the conditional ‘if forms
should break away … etc.’ Abelard discusses at length what he calls a ‘terrible
heresy’ which derives from this position (i.e. the identification of the substance
of anythings with divine substance).
The next criticism concerns the thesis that universals are individual things
considered from a certain point of view (the theory of non-difference).
According to this stance, things are not discrete only by their accidental prop-
erties, but also in their essence.29 This means that, even if it were possible
to isolate a substantial part, this constituent would be linked to the thing in
which it is ‘immersed.’ Abelard highlights the following point:
28 Log. Nostr., 515, 32–36: ‘si hanc sententiam concedi conuenit, quippe si formas contingeret
a subiecta materia discedere, ita scilicet quod subiecta his penitus carerent, in nullo peni-
tus hic et ille differrent, sed iste et ille omnino idem efficerentur.’
29 Log. Nostr., 518, 21–27: ‘Dicunt enim singulas substantias ita in propriae suae essentiae dis-
cretione diuersas esse, ut nullo modo haec substantia sit eadem cum illa, etiamsi substan-
tiae materia penitus formis careret, quod tale secundum illos ‘praedicari de pluribus’ ac
si dicatur: Aliquis status est, participatione cuius multae sunt conuenientes; ‘praedicari
de uno solo’ tale est ac si dicatur: Aliquis status est participatione cuius multae sunt non
conuenientes.’
The Position of Abelard on Porphyry 217
individual is not predicated of more than one thing. This clarification, how-
ever, does not resolve the problem at hand. Abelard claims he is amazed by the
error ‘tanto numerosae multitudinis’, and thereby implicitly provides informa-
tion about the diffusion of the identity thesis.
The second criticism concerns the interpretation of the sentence ‘homo
ambulat’. According to Boethius, the sentence is false if refers to the species
(de homine speciali). Indeed, how can it be true of the particular man if the
universal is equal to the particular? The predictable answer that a particular
man, as universal, does not walk, can be understood in two different ways:
Thus, if the reason why the sentence ‘(the special) man walks’ is false is (b),
the sentence would be false of the particular for the same reason. This would
be contrary to Boethius’ thesis, and we may add, contrary to common sense.
The next criticism is supported by a Boethian text that argues that the uni-
versal species/genera are predicated in quid of subordinate things. We know
that this has to do with the conditions that restrict the relation of predication
(the narrow interpretation of ‘says-of’ is obtained by attaching an adverb like
‘essentially’ to the right-hand side of principles P: see Chapter 1). There are,
of course, margins of uncertainty but in general this position is rather clear.
Abelard now objects to the identity theorists who argue that in certain cases
the condition imposed by Boethius does not apply:
30 Log. Nostr., 520, 10–14: ‘Quid enim est quod ait Boethius […] cum secundum eos indi-
vidualitas sit quaedam forma quae in omnibus rebus individualibus accidentaliter inest?
Itaque et huic individualitati per accidens inest, cum secundum hoc individualitas
The Position of Abelard on Porphyry 219
Abelard considers a rather peculiar case here: the individual is not the usual
first substance (like Socrates), but a quality, this individuality. The identity the-
ory postulates the presence of the properties of universality and individuality
in the individual; accordingly, these properties are not necessary but acciden-
tal (i.e., they may or may not occur at a certain time, depending on how the
individual is considered).31 Thus, individuality is predicated accidentally of
this individuality, but at the same time is predicated essentially of it, since it is
inconceivable that this individuality is not an individuality.
According to identity theorists, to say that a thing is predicated means that
it is predicated essentially or accidentally. Abelard tries to show that from this
position it follows the wrong consequence that the concept of predication is
ambiguous. Besides general difficulties, which concerns the tightness of topi-
cal laws, there is a more specific one: this whiteness (which is something in-
dividual) is predicated in essence of itself (as a noun) and in adjacency32 of a
given individual (via an adjective in predicate position). In this way, this partic-
ular whiteness would be a universal, being predicated (in any sense) of several
things against the common doctrine and common sense.
The latter objection concludes the section devoted to the theory of identity.
The different ways in which Abelard presents his criticism, and his use of his
entire repertoire of dialectical subtleties, suggest that he considered the theory
absolutely worthy of attention. The following position considers the univer-
sality as a property attributed to particular things. The thesis, as usual, is for-
mulated rather enigmatically, and there are several possible interpretations. It
seems that the first Boethian view on universals is at stake:
There are those who seem to say that there are universal things. In trying
to explain how those things are common, they say that a thing can be
universal or particular, i.e. it has a certain property, as a universal, and
another property as a particular. In the same way, /animal/ and /body/
individuum dicitur, quomodo et Socrates. Nec tamen minus individualitas de hac indi-
vidualitate praedicatur in quid.’
31 It is not entirely clear to me whether this is a true accidental nature or some epistemic
property. This depends on how we want to interpret the fact that one thing is universal or
particular when observed from different points of view.
32 It is difficult to briefly explain what exactly adjacency is, since one would have to take into
account the logical and ontological aspects, as well as relate these to the theory of truth.
In short, we can say that an expression is predicated in essence when it means an essen-
tial component of its reference (in context). For example, ‘white’ means adjacent white-
ness, because in ordinary predicative contexts it refers to white things; but it is predicated
in essence (meaning whiteness) in sentences like ‘this color is white’.
220 Chapter 8
This text contains something that comes close to the formulation of the
Aristotelian principle of predication. A first observation is that if we accept
that x is universal if and only if there are several things any one of which is
an x, then it is also true that the individual is universal iff there are several
things each of which is an individual. Could this mean that proponents of this
thesis share the view criticized above, according to which the same particular
things are universal from a certain point of view? I do not think so; it is more
likely that Abelard wants to highlight the difficulties arising from the proposi-
tion ‘the individual is a universal’. The following is one way to reconstruct this
argument:
In general it is not true that:
– The individual is universal iff there are several things each of which is an
individual.
The latter is false because the individual is not a universal, while there are sev-
eral individual things. Thus, the general principle is not true, since it allows at
least one exception.
The criticism does not appear to be very convincing; the premise is that
/man/ is a universal. For the same reason /individuality/ and /individual/
must also be considered universals. Even more troubling is the conundrum
that arises if we assume the predicate ‘is non-universal’: if there are several
33 Log. Nostr., 521, 21–29: ‘Sunt iterum alii qui aliter res universales esse viedentur affirmare.
Hi rebus communitatem assignantes dicunt rem aliam universalem, aliam singularem,
id est alterius proprietatis esse ex hoc quod est universalis, et alterius ex hoc quod est
singualeris. Ut animal est universale et corpus, nec tamen aliquod animal vel aliquod
corpus. Tale est secundum hanc sententiam animal esse universale, ac si dicatur: plura
sunt quorum unumquodque animal est, et tale est hoc anima praedicari de uno solo, ac si
dicatur: una sola res est hoc animal.’
The Position of Abelard on Porphyry 221
things that are not universal, then /non-universal/ is universal.34 We see here
two problems related to the interpretation of the principle of predication. One
concerns generic entities, which Boethius had already considered poor candi-
dates to fill the role of universals; the second has to do with the logical difficul-
ties that accompany the history of the Aristotelian principle of predication.
The next thesis – which Abelard judges to be ‘more reasonable’ (rationi
vicinior) – assigns universality to meaningful expressions (sermones). As we
have seen, several Aristotelian and Boethian texts support this interpretation.
Abelard considers the arbitrariness of the attribution of meaning:
So, vocal sounds, as things (i.e. phonic material), are not universal. They be-
come universal only through a process of imposition, a sort of baptismal act in
which the vox is charged with a meaning and becomes sermo. Admittedly, this
does not seem to be the crux of Abelard’s criticism. In order to clarify this last
thesis, let us examine the path that the analysis follows:
– Question: why is it that sermones can be universal and voces cannot, since
sermones are voces?
– The difficulty is resolved in this way …
– There are those who want to rephrase the same difficulty in other terms
– The (reformulated) difficulty is resolved in this way …
– There follow clarifications about the allowed and rejected uses of terms like
‘species’, ‘genus’, ‘vox’, ‘sermo’
The criticisms put forth are invalidated by Abelard himself (in one case he ac-
tually calls some of the objections ‘frivolous’). I suspect that this approach is
34 Given the predication principle stated in the first chapter: X℘’a iff a is an x (X is said of a
if and only if a is an x) and that X is a universal (as a species).
35 Log. Nostr., 522, 28–32 : ‘Sic ergo sermones uniuersales esse dicimus, cum ex natiuitate,
id est ex hominum institutione, praedicari de pluribus habeant; uoces uero siue res nul-
latenus uniuersales esse, etsi omnes sermones uoces esse constat. Si enim aliqua res de
pluribus praedicaretur, utique eadem in pluribus reperiretur.’
222 Chapter 8
We admit the truth of these sentences […] and similarly of all sentences
in which the subject is a vocal sound involving a conventional process
and notifying a community of elements, not just an essence or the same
utterance (like genus, expression, word, vocal sound, utterance) but a sig-
nification. A vocal sound that simply points to an essence is of the type:
animal, speech sound, sound.36
36 Log. Nostr., 524, 6–10: ‘concedimus itaque has esse veras […] et similiter omnes in quibus
subicitur uox innuens institutionem, non simpliciter essentiam uel prolationem, sed sig-
nificationem et praedicans communitatem, sicut est: genus, uniuersale, sermo, uocabu-
lum, dictio, oratio. Vox autem simpliciter innuens essentiam est ut animal homo, uox,
sonus aeris etc.’
37 Log. Nostr., 522, 30; 524, 3–17.
The Position of Abelard on Porphyry 223
homo est hic status animal hic status animal est <homo>
Socrates est materia Socratis haec materia Socratis est Socrates
aliquid est utrumque istorum utrumque istorum est aliquid
In the upper list, vox (speech sound), as a subject, refers to the material com-
ponent of a linguistic sign, of which ‘species’, ‘genus’, and ‘universal’ cannot
be predicated. The rejection of the examples having ‘this (a) speech sound’ as
the subject requires some deviation from the canonical Aristotelian-Boethian
doctrine. In fact, if we assume that the sermo is a meaningful vox, it is quite
legitimate to use the genus name in a phrase to denote an individual: indeed
‘this animal is (a)(this) man’ has the same form as the sentences ‘this vox (re-
ferring to a sermo which is a kind of vox) is a genus’ and ‘this vox is this sermo’.
This deviation, though strained, seems to have been consciously introduced by
Abelard to prevent unwanted interpretations of his theory of sermo.
Some literature exists on the formulas in the second list.38 I have argued else-
where39 that the verb ‘to be’ might mean a metaphysical relation between the
species and the genus. This relation is non-reflexive, asymmetric and transitive.
What, then, makes these formulas acceptable? Abelard speaks of expressions
that signify things (i.e. they are used), not expressions of the metalanguage. He
does not offer much more; since the text under discussion is part of (or should
be part of) a broader argument, it contains no solutions but only clarifications.
It is the task of the pars construens, in which Abelard introduces his position
on the Porphyrian issue, to explain why propositions like ‘This speech sound
(vox) is universal’ are acceptable.
The final part of the chapter is devoted to the exposition of the author’s
views on the Porphyrian problem. After briefly mentioning Boethius’ position
on the effectiveness of universal concepts, Abelard resumes the interrupted
discussion from the previous section on the acceptability of statements like:
/animal/ is a genus
/animal/ is in several things
/man/ is universal
/man/ is a species
Abelard essentially says that only those who do not understand the use-
mention distinction.40 can claim that ‘man’ and ‘animal’ have an ordinary
meaning in these sentences. Anyone familiar with the technicalities of dialec-
tics should be able to interpret the questions asked by Porphyry, and to distin-
guish between the literal meaning and the intended meaning of the sentences
in which terms of ‘second intention’ occur. The text which follows illustrates
this approach:
There is another way to solve the problem: genera and species subsist,
i.e. refer to subsistent things, of which they are nouns or participles
(nomina sicut et participia). The same thing is part of the extension
(continetur) of the universal and particular noun, and in this passage the
verb ‘subsist’ – by adjoining the nouns ‘species’ and ‘genus’ that mean
linguistic expressions – is used in a figurative sense to signify a linguistic
expression and not something real. Just like the noun ‘substance’, which
properly befits things, is sometimes used to mean nouns.41
40 Obviously, there is much plaguarizing of Abelard’s assertion. I limit myself to quoting
just one contemporary philosopher, who states that ‘many philosophical works have been
vitiated by a failure to distinguish use and mention’ (M. Devitt – K. Sterelny, Language and
reality: an introduction to the philosophy of language, Blackwell, Oxford 1999, p. 40).
41 Log. Nostr., 525, 30 – 526, 2: ‘soluens uero aliter hoc modo scilicet: genera et species sub-
sistant, id est res subsistentes appellent, uidelicet eas quarum sunt nomina sicut et parti-
cipia. Eadem namque res ab uniuersali nomine et particulari continentur et hoc loco hoc
uerbum ‘subsistit’ de rebus ad sermonem transfertur per adiunctionem horum nominum:
genus et species, quae sermonibus data sunt. Veluti substantia hoc nomen quod proprie
rebus conuenit, tamen in designatione nominum quando large accipitur.’
The Position of Abelard on Porphyry 225
Universals are only concepts i.e. universals mean things without perceived
formsa
Universals are corporeal i.e universals denote particular thingsb
Universals are incorporeal i.e. universals denote particular things in a
common wayc
Universals are in what is perceived i.e. universals denote objects of perceptiond
by the senses
Universals are beyond what is i.e. mean things without some form which is
perceived by the senses subject to senses … are beyond perceived
things by their significatione
a Log. Nostr., 526, 12–13: ‘… significantes rem non cum aliqua forma quae sensui subiaceat.’
b Log. Nostr., 526, 31: ‘sunt corporalia, id est res discretas nominant.’
c Ibid., 39–40: ‘et sunt incorporalia, id est res discretas indiscrete nominant’.
d Log. Nostr., 527, 23–24: ‘in sensibilibus sunt posita, hoc est sensibilia habent nominare.’
e Ibid., 25–29: ‘ponuntur extra sensibilia, id est res habent significare et non cum aliqua forma
quae sensui subiaceat … sunt extra vero per significationem.’
Here Abelard follows the path laid in the Ingredientibus, and continues unhin-
dered until he arrives back at the place where he began. In fact, the unanswered
question about the ontological status of universals is reformulated on seman-
tic grounds. Abelard obviously shares the point of view of those who argue that
there are only perceptible things (in addition to what is believed by faith).42
On the other hand, he also argues that universal terms mean something; he
42 Abelard’s position on this issue is actually quite elusive, as he claims that things said about
these objects are enveloped in riddles, parables and metaphorical uses of expressions.
See, for example, Theologia Christiana, ed. E.M. Buytaert, Brepols, Turnhout 1969, 245:
226 Chapter 8
‘Unde in Deo nullum propriam inventionem vocabulum servare videtur, sed omnia quae
de Eo dicuntur, translationibus et parabolicis aenigmatibus involuta sunt.’
43 Log. Nostr., 530, 24–31 : ‘Solet enim quaeri de significatione et intellectu istorum (id
est uniuersalium nominum) quas res scilicet significare habeant. Nam cum audio hoc
nomen ‘homo’ quod pluribus commune est rebus ad quas aequaliter se habet, quam rem
in ipso intelligam quaero. Quodsi sicut oportet, respondebit, quod homo ipso intelligatur,
sequitur autem quaestio, quomodo hoc uerum sit, nisi hic homo uel alius intelligatur,
cum scilicet omnis homo sit hic uel ille uel alius.’
44 Log. Nostr., 531, 5–6: ‘sed omnis intellectus intelligens hominem est intelligens hunc uel
illum.’
The Position of Abelard on Porphyry 227
answer (for example) ‘I want a gold castle.’ In this case, the object of which we
speak is a gold castle; the something which the general and specific terms talk
about seems to be made of the same material as the imaginary castle.
The topic of non-denoting terms has a long history; we shall limit ourselves
to recalling the discussion which arises from some of Aristotle’s remarks in
the De Interpretatione (for instance, on ‘second adjacent’ sentences), the ob-
servations of Neoplatonic commentators on mental fictitious entities, the
Carolingian treatise De Nihilo et Tenebris, and, closer to Abelard, St. Anselm’s
De Casu diabuli. If we consider St. Anselm’s work, we must ask how it is that
a sentence like ‘the nothing [evil] is F’ could be true if the subject does not
denote anything. After trying to convince the Disciple that ‘nothing’ means a
privation, at a certain point the master of De Casu diabuli, forced by the inter-
locutor’s insistence, seems to accept that it does have a meaning sui generis.45
Abelard ends the analysis of ‘intensional’ contexts by proposing to consider
equivalent the sentences ‘I want x’ and ‘I have the will to have x’. However, this
cannot be considered sufficient, as the gold castle appears to be different from
the desire to have it, who wants a castle would not settle for a willingness to
have it instead of actually having it. In fact, Abelard does not insist on shifting
from the object wanted to the will wanting it; in his comments on the example
above, ‘Chimera intelligo’, he states that ‘in spite of the fact that the thing of
which I am thinking does not exist, I am still thinking of something.’46
45 Similarly, Abelard admits that terms like ‘chimera’ can have an impositive meaning: ‘sicut
enim dictum est in impositione hominis, dicatur ista res homo, sic etiam dictum est in
huiusmodi re non existente: dicatur ista res opinabilis, ut sic etiam non existentium no-
mina inventa sint ad agendum de ipsis secundum hoc quod in impositionem veniunt.’
(Dialectica = Dial., ed. L.M. de Rijk, Van Gorcum, Assen 19702. 137, 32–35) The Stoic theory
of meaning may come to mind: the lekton is what can be said. For the Stoics, it is not a
thing in the strict sense, but an entity belonging to a third realm, beyond that of things and
of linguistic expressions (cf. S. Bobzien, ‘Stoic Logic’ in K. Algra – J. Barnes – J. Mansfeld –
M. schofield (edd.), The Cambridge History of Hellenistic Philosophy, Cambridge University
Press, Cambridge 1999, 92–157). This similarity may, however, be merely coincidental,
since the connection between Diogenes Laertius, Sextus Empiricus and the twelfth cen-
tury is tenuous at best.
46 Tweedale thinks of a non-denotative use of terms, and of referentially opaque contexts
(see Abailard on Universals, quot., 167–168, 180–184). There are at least three drawbacks
to this approach: a) there is no adequate textual evidence of an analysis of natural lan-
guage in terms of ‘logical’ language (as the author himself acknowledges in op. cit.,
‘Introduction,’ p. 9); b) the conditions of truth recognized by Abelard require that the
terms of a categorical statement denote something (existing or not existing); c) to keep
(b) Abelard admits – albeit hesitantly – that there may be an imposition of nouns on ficti-
tious entities.
228 Chapter 8
Abelard tries in his commentaries to find a way to combine the ideas that
universal meanings are something and that they cannot be counted among
existent objects. The text considered here is no exception, and may be the best
example of this seemingly impossible mission. If Abelard is a nominalist, he
is certainly a strange one;47 indeed, in analysing the examples quoted above,
some modern philosophers point to the misleading nature of natural language.
Those who see things from a different perspective, however, may find Abelard’s
suggestions interesting and worth developing.
47 In fact, any number of labels have been applied to Abelard’s philosophy (an almost com-
plete list was made by R. Corazzon, downloadable at: www.ontology.co/biblio/Abelard
-biblio-one.htm and at: www.ontology.co/biblio/Abelard-biblio-two.htm. For another
list, which would certainly be longer if it were updated, see C. Wenin, ‘La signification
des universaux chez Abélard’, Revue philosophique de Louvain, 80, 47 (1982), 414–448, cf.
especially pp. 417–19). The most neutral of these is ‘non-realism’ or ‘unrealism’ (P. King,
‘The Metaphysics of Peter Abelard’ in The Cambridge Companion to Peter Abelard, cit.,
65–125). An extreme example (though not the worst) is Maloney’s assertion that Abelard
presents a ‘hybrid of nominalism and conceptualism with results definitely platonic’
(JC Maloney, ‘Abelard’s Theory of Universals,’ Notre Dame Journal of Formal Logic, 23, n. 1
(1982 ), 27–38). Tweedale, op. cit., 8–10, makes some considerations on this subject. Labels
aside, one must at least identify something to which they may be attached. I do not think
that there is sufficient clarity in the distinction between ‘natural philosophy’ (precisely
that part of Abelard’s philosophy in which he develops, or apparently struggles to develop
some sort of hylomorphism) and semantic theory; in other words, between the part of his
research about existing things and the one that is (at some level) involved in an ontology
of objects, in the general sense of the term.
The Position of Abelard on Porphyry 229
P1Abelard.c ‘X’ ℘ ‘x’ if and only if the meaning of ‘x’- the meaning
(status) of ‘X’ is a fact
P1Abelard.d the meaning of ‘x’- the meaning (status) of ‘X’ is a fact if and
only if x is a(n) X
Let us disregard the hyphen for a moment and look at the possible candidates
for the role of universal: one is represented by the term ‘X’, another is the mean-
ing of the same term, the status. ‘Universal’ is, by definition, what is predicated
of several things. One would expect Abelard to share a broad interpretation of
the concept of universal, such that it includes participles (verbs can be turned
into the form ‘is + participle’) and adjectives. In the Logica Ingredientibus, in
fact, there are confirmations of this, along with some clarifications.
Commenting on the definition of genus, Abelard considers the conditions
imposed by Porphyry and Boethius on predication:
The third condition clearly excludes adjectives, not because they cannot be
predicated,49 but due to the restrictions imposed upon the Porphyrian tree.
48 Log. Ing., 35, 33 – 36, 3: ‘Tria itaque ponit, per quae generale uocabulum ab omnibus aliis
diuidit, scilicet (i) praedicari de pluribus, quod est esse uniuersale, per quod a singulari-
bus, id est diuersis, separatur; et (ii) praedicari de differentibus specie, hoc est de no-
minibus res tantum diuersarum specierum naturaliter continentibus, per quod differt
a specialissimis et eorum propriis; et (iii) praedicari <in> eo quod quid sit, id est reddi
conuenienter ad interrogationem factam per ‘quid’, non per ‘quis’ uel ‘quae’ uel ‘quod’,
quae magis propria nomina requirunt, uel per ‘quale’ uel ‘quomodo se habet’, quae sump-
ta tantum uocabula exigunt, ut ‘rationale’, ‘album’ ‘sedet’, ‘iacet’.’
49 Adjectives as well as common nouns can play a predicate function; Porphyry expressly
states that among things that can be predicated there are species and genera as well as
230 Chapter 8
differences, essential properties and accidents. Abelard clearly acknowledges that ad-
jectives, nouns and participles can be predicated or are inherent (in a general sense of
the term) in a subject. He is somewhat hesitant, however, due to a conflict between the
terminology of being-in and that of inherence: ‘quod ait Boethius dici de subiecto idem
esse quod uniuersale […] et rursus esse in subiecto idem esse quod accidens […] Quippe
Aristoteles in definitione uniuersalis, quam in Peri ermeneias ponit, quae est: ‘aptum
natum praedicari de pluribus’, consentire uidetur, ut album quoque, quod sumptum est,
uniuersale sit, quod tamen ipse in tractatu substantiae de subiecto dici non concedit’
(Log. Ing., 132, 18–26).
50 I believe Frege’s notion of Begriff can be used as explicans for the notion of status (mean-
ing of a predicate; cf. Dial., ed.cit., 111–112). Some studies have drawn attention to the pos-
sible use of Fregean notions in clarifying Abelard’s theoretical positions: cf. M. Tweedale,
Abelard on Universals, North Holland, Amsterdam 1976 (see in particular the observa-
tions contained in Section 6); K. Jacobi ‘Abelard and Frege: the Semantics of Words and
Propositions’, in Atti del convegno internazionale di storia della logica, S. Gimignano, 4–8
dicembre 1982, Clueb, Bologna 1983, 81–96; P. King, ‘Abelard’s answers to Porphyry,’ cit.
(these papers try to find notions in the Medieval texts corresponding to the modern no-
tions of sense and denotation). I believe that the comparison between Abelard and Frege
has its own utility; in particular, I suggest examining the following pairs of concepts: sta-
tus – Begriff, intellectus/sensus – Sinn, denotatio, nominatio – Bedeutung.
The Position of Abelard on Porphyry 231
are dealing with here, which concerns the nature of things and the truth
of their status.51
When we say ‘running man’ or ‘white man’ […] we mean a certain inher-
ence of running or whiteness in man; we refer without a doubt to the
adjacency of the accident to the foundation [i.e., the meaning of the
subject].52
This relationship of inherence is described at both the syntactic and the seman-
tic level. In the case of the definition, Abelard speaks of a copulatio-(ad)iunctio
of the differences to the genus, and of a syntactical composition that ‘mimics
51 Log. Ing., 17, 12–21: ‘Nota autem aliam esse coniunctionem constructionis quam attend-
unt grammatici, aliam praedicationis quam considerant dialectici […] Coniunctio itaque
constructionis totiens bona est, <quotiens> perfectam demonstrat sententiam, sive ita sit
sive non. Praedicationis vero coniunctio quam hic accipimus ad rerum naturam pertinet
et ad veritatem status earum demonstrandam.’
52 Dial., 149, 5–11: ‘Cum dicimus ‘homo currens’, ‘homo albus’ […] quaedam insuper inhae-
rentia cursus vel albedinis ad hominem in ea exprimitur […] Cum enim cursum vel al-
bedinem circa hominem per ‘currens’ vel ‘album’ determinamus, quandam procul dubio
cohaerentiam accidentis ad fundamentum secundum adiacentiam innuimus.’ On adja-
cency intended as real relation, see: Log. Ing. 260.
232 Chapter 8
the real one of matter and form.’53 In all cases of grammatical composition, we
have an expression that has the additional capability of linking to other expres-
sions in order to give rise to a meaningful compound expression. A standard
categorical statement is not an exception from this point of view: the predicate
expresses a meaning (status) and a connection that has the form of the copula
on the syntactical level. Thus, the standard Latin categorical sentence with a
singular subject is analyzed in the following way:
[Socrates [est[homo]]]
Abelard points out that inherence is not a real property (in fact, he calls it a
proprietas nulla). If it were, the following difficulties would arise: i) if ‘is’ meant
the relation of inherence, such a relation would act as predicate and the con-
stituent terms of the sentence would become its arguments; ii) if the predicate
in ‘Socrates est animal’ meant being animal (by its nominal part) and inherence
(on the copulative side), then the same statement would predicate two dif-
ferent properties and therefore be equivocal; iii) if the main predication con-
cerned inherence, the term ‘animal’ would no longer be predicate but subject
(a rewording of the statement would be ‘the animal is inherent in Socrates’).
Point (iii) is particularly interesting, because it is an implicit criticism of a ver-
sion of the principle of predication:
A is inherent in x iff x is an A,
so
56 On the syntactic analysis of the sentence in two parts, verbal and nominal, see the refer-
ences provided by K. Jacobi in ‘Philosophy of Language,’ cap. 4 di J.E. Brower – K. Guilfoy,
The Cambridge Companion to Abelard, cit., p. 148 e nota 93.
234 Chapter 8
slightly modified, could also be taken as his conclusion: ‘the difficulties of the
text are not [only] <so much> the ones implicit in Porphyry’s questions about
genera and species; other no less challenging ones [may be added] <should
rather be considered>, such as the problem of meaning and common cause of
the imposition.’57
Gilbert of Poitiers
When we say ‘a body is’ or ‘a man is’ and so on, theologians understand
‘being’ to be said by a certain extrinsic derivation from the essence of
its principle. They do not say a body is by reason of its corporeality, but
that it is something; nor that man is by reason of humanity, but that it is
1 Among introductory studies one can consult the following: L.M. de Rijk, ‘Gilbert de Poitiers,
ses vues sémantiques et métaphysiques,’ in J. Jolivet – A. de Libera (edd.), Gilbert de Poitiers
et ses contemporains. Aux origines de la Logica Modernorum, Bibliopolis, Napoli 1987, 147–171;
J. Marenbon, ‘Gilbert of Poitiers,’ in P. Dronke (ed.), A History of Twelfth-Century Western
Philosophy, Cambridge University Press, Cambridge 1988, 328–352; J. Marenbon, ‘Gilbert of
Poitiers’, Chapter 35 in J.E. Gracia – T.B. Noone, A Companion to Philosophy in the Middle Ages,
Blackwell, Malden 2005.
We immediately notice the recurring use of the ablatives ‘corporeitate’ and ‘hu-
manitate’ throughout the text (expressions like ‘humanitas’ and ‘esse homo’ are
here translated with ‘being man’; in general the name of a subsistence will be
‘being x’, where ‘x’ stands for a noun or an adjective).3 The formula ‘x is not by
reason of being x’ is used to exclude the possibility that a world entity exists
because of something other than its principle, i.e. God. The text quoted also
excludes that there is a being without qualifications.
In the faculties ‘where genera are considered things which draw their being
from the principle’,4 we find several different interpretations of the second
Boethian axiom:
Some say that the themes, namely materials, which sentences are about,
are and are something for the same reason. They say, therefore, that the
verb ‘to be’ is equivocally predicated of everything.5
Others draw a distinction: they say those things that subsist are by rea-
son of the subsistences, and are something by reason of those things con-
nected with the subsistences.6
2 Expositio in Boecii librum De Bonorum Ebdomade (=In De Bon. Ebd.), in N.M. Häring (ed.),
The Commentaries on Boethius by Gilbert of Poitiers, Pontifical Institute of Medieval Studies,
Toronto, 1966; 193, 55–62: ‘Cum enim dicimus ‘corpus est’ vel ‘homo est’ vel huiusmodi, theo-
logici hoc esse dictum intelligunt quadam extrinseca denominatione ab essentia sui princi-
pii. Non enim dicunt corporalitate corpus esse sed esse aliquid: nec humanitate hominem
esse, sed esse aliquid. Et similiter unumquodque subsistens essentia sui principii predicant
non esse aliquid sed esse: illa vero que in ipso creata est, subsistentia non esse sed esse
aliquid.’
3 English has abstract forms which correspond to Latin, e.g. ‘corporalitas’ vs ‘corporeality’, but
in such cases the correspondence with the concrete form is sometimes lost in translation
(e.g., corpus- corporalitas vs body-corporeality). For this reason I will translate the Latin -xtas
form throughout as ‘being x’. In translating Gilbert’s forms some authors use the suffix ‘ness’,
for instance ‘stone-ness’, ‘Plato-ness’, etc. However, this makes it difficult to preserve the uni-
formity of Latin technical usage.
4 In De Bon. Ebd. 193, 29, 67–68.
5 In De Bon. Ebd. 193, 67–70: ‘alii quelibet illa orationum suarum themata i.e. materias, de qui-
bus loquuntur, eodem quo dicunt esse dicunt etiam esse aliquid. Unde etiam hoc verbum
‘est’ dicunt de omnibus equivoca predicari.’
6 In De Bon. Ebd. 194, 71–72: ‘Alii vero dividunt. Et ea que subsistunt dicunt esse subsistentiis et
esse aliquid his que subsistentias comitantur.’
Gilbert of Poitiers 237
Our author claims he considers both points of view; however, except for the
seventh Boethian axiom, he simply gives an exposition of the text in naturali-
bus. Indeed, in the following comment the notion of being is identified with
that of subsistence:
Boethius claims, therefore, ‘diversum est esse’ i.e. the subsistence which
is in a subsistent, ‘and id quod est’, i.e. the subsistent in which the subsis-
tence is: for instance being body and the body, being man and the man.9
Gilbert usually uses a formula like ‘x is a y by reason of being y’; in other words,
the subsistence is that by which something is a certain thing. What Gilbert
says about the ‘being in’ of a subsistence suggests that he is referring to a rela-
tionship between particular and universal. To understand his point of view on
universals and the relation of inherence, however, we must proceed with care
in order to avoid misunderstanding. In a passage following the one just quoted,
he adds that the intellect ‘somehow’ separates subsistence from that in which
7 I believe it is still the opinion of members of the faculty of theology as the passage quoted is
followed by a summary of the position of these philosophers.
8 In De Bon. Ebd. 194, 81–83: ‘Sed secundum alios philosophos’ esse’ subsistentium sole illorum
que predicantur subsistentie: ‘que vero sunt’ ea tantum que illas in se habendo subsistunt.’
9 In De Bon. Ebd. 194, 90–92: ‘Ait ergo: ‘diversum est esse’ i.e. subsistentia, que est in subsistente,
‘et id quod est’, i.e. subsistens in quo est subsistentia: ut corporalitas et corpus, humanitas
et homo.’
238 Chapter 9
it is, and this will have to serve, at least temporarily, to quell some of our fears
about being-in.
The distinction between ‘being body’ and ‘the body’ and ‘being man’ and
‘the man’ should not necessarily be taken to indicate a generic body or man.
‘Being a body and the body’ could mean ‘being a body and the particular thing
in which the subsistence of being a body inheres’. Again, it should be noted
that there is no explicit difference between a possible referent of common
nouns (e.g. ‘homo’), intended in suppositio simplex, and its being (e.g. humani-
tas). This fact is of some importance. The examples presented provide us with
the tools to work out what subsistences and subsistents are: the former are seen
as being-x and the latter are the individuals referred to by ‘x’. As we will see, it
is not an accident that a noun like ‘body’ is used as an example: being x is not a
subsistence for every x, only substantial being-x (being man, being animal..) or
essential properties (being rational) can be properly called subsistences.
When commenting on the fourth Boethian axiom, ‘diversum est tantum esse
aliquid et esse aliquid in eo quod est’’, Gilbert states that subsisting things are for
subsistences only, and are something in virtue of accidents. Boethius claims
that ‘esse aliquid in eo quod est’ refers to the substance:
When ‘it is something’ is said of a thing, in such a way that one can add
‘in eo quod est’, the substance is meant, i.e. the subsistence, that for which
it is said ‘to be something’ in such a way.10
10 In De Bon. Ebd. 198, 84–86 : ‘Cum de quolibet dicitur ‘est aliquid’, ita quod potest addi ‘in eo
quod est’, significatur esse substantia, hoc est subsistentia, id quo ita ‘esse aliquid’ dicitur.’
Gilbert of Poitiers 239
the men, stones, white things, rational things, bodies, Plato quod ests; the
humanity, stone-ness, whiteness, rationality, corporeality and Plato-ness
quo ests.11
It is to be noted that Marenbon associates quo est with the meaning of any ab-
stract noun derived from a common noun or adjective. We have encountered
examples of these expressions, but the question is what kind of relationship
subsists between a quod (signified by a noun-adjective) and a quo (signified
by the corresponding abstract form). Gilbert does not explicitly mention the
pair id quod-id quo, as he does for id quod-esse and subsistens-subsistentia. The
concept of id quod is evidently the same as that which appears in Boethius’
second axiom; that of id quo is to be sought and interpreted in the context in
which it appears. In fact, the expression ‘id quo’ as opposed to ‘id quod’ occurs
in few places, one of which is the following:
While there are relatively few other occurrences of id quo, there are numer-
ous instances of the ablative of the relative pronoun in contexts such as the
following:
Whether or not this list is complete, the search for further occurrences becomes
rather pedantic when one recognizes, as De Rijk does, that esse = subsistentia =
id quo and subsistens = id quod. 18 Subsistences are that for which a subsistent
is something. This being something appears to express modality. If I am not
mistaken, it is only in two cases that non-essential qualitative and quantitative
properties take the place of subsistences. There are, however, other texts in
which Gilbert, despite not using the ablative of the relative pronoun, discusses
the causal powers of essential and accidental properties.19 In his commentary
on the De Trinitate the humanitas, together with its essential components, is
presented as something which determines man’s being. Subsequently, Gilbert
adds that accidents also cause a subsistent to be a qualified thing.
The following is a more general text which seems to equate the causality of
subsistences and accidents:
15 I quote from the Commentaries on Boethius, disregarding the reference to individual com-
mentaries: 58, 46; 77, 81; 81, 80–81; 118, 1–3; 135, 93–94 e 97–100; 144, 79–80 (the subsistent
is by reason of the subsistence); 258, 30–31; 262, 39–40; 319, 59–60 (where Gilbert speaks
of nature as substantial form, as well as subsistence).
16 Commentaries, 135, 3–4.
17 Commentaries, 136, 7–8.
18 L.M. De Rijk, ‘Semantics and Metaphysics in Gilbert of Poitiers. A Chapter of Twelfth
Century Platonism,’ Vivarium XXVI, 2 (1988), 76.
19 Marenbon recognizes that Gilbert does not always use the pair ‘id quod’ – ‘quo est’, but
considers the different terms used (‘subsistence’, ‘to exist’, ‘form’, ‘substance’, ‘subject’) as
expressing the same distinction.
20 In Contra Euticen, 245, 77–81: ‘Nativa namque per aliquam sui vel efficientem vel effici-
endi proprietatem concipiuntur: ut album per albedinem et albedo per naturam faciendi
Gilbert of Poitiers 241
So, from the point of view of doing, accidents like whiteness are compared to
essential properties such as being man or being rational.21 Gilbert expresses
himself in similar terms in the commentary to De Trinitate:
That being the case, it should be added that every subsistent subsists for
the concretion of many things for which it is, i.e. genus, difference and
accident, and for this reason a thing is for this something and for some-
thing else.22
In conclusion, we can say that Gilbert considers any abstract meaning to have
causal efficacy, in the sense that it enters the metaphysical composition of a
subsisting thing in some form and according to some modality, and also in the
sense that a thing is said or conceived according to the meaning of the cor-
responding predicative expression, e.g. ‘man’ compared to ‘being man’, ‘white’
compared to ‘being white’.
At this point, after following a relatively level path, the road begins to climb
and some roughness can be detected: the concept of causality, the ontological
status of subsistences, predication, the relationship between Aristotle’s catego-
ries and the general concepts introduced by Boethius in his theological writ-
ings. In the next paragraphs we will attempt to explore each of these aspects.
In the history of creation told by the philosophers of the Chartres school there
exists a certain time sequence, with intermediate stages between the creation
of matter and the formation of particular things. For example, according to
William of Conches created matter is an original confusion of ‘simple and very
small particles’; the four elements (fire, air, earth, water) originate from matter,
and singular existing things stem from these substances. According to Gilbert,
there are two senses of primordial matter:
album. Nichil enim naturalium nisi per causam et nichil mathematicorum nisi per effici-
endi potestatem concipi potest.’
21 Subsistences are definitely among mathematical things, in the sense in which Gilbert
uses the term.
22 In De Trinitate, 144, 79–81: ‘His ita se habentibus, addendum est quod omne subsistens
multorum quibus est – i.e. generis et differentie et accidentis – concretione subsistit, ac
per hoc aliquo alioque est aliquid.’
242 Chapter 9
Not only ylen, which Plato calls ‘silva’ – which ‘according to the philoso-
phers’ is without being something – but also what is not said to be because
of its noun, but is said to be something due to its perpetual subsistence.23
This explanation leaves several points unclear. The raw material (ylen) is dif-
ficult to understand: it is an abstract entity, but in a very special sense, since
it was created formless. At the same time, it does not appear to be a form but
a kind of support to properties. Gilbert speaks of it as something ‘incompre-
hensible, but intelligible.’ Equally problematic is the concept of ‘pure sub-
stances’ understood as something distinct from that of ‘material elements.’
Gilbert claims that the latter are in the silva, whereas their formal counterparts
23 In Contra Euticen, 335, 23–27: ‘Non modo ylen, quam Plato ‘silvam’ nominat – que quidem
secundum philosophos est sed non est aliquid – verum etiam illam que non suo nomine
dicitur esse sed secundum suam perpetuam subsistentiam dicitur aliquid esse.’
Gilbert of Poitiers 243
are derived from silva ‘for an intelligible species.’ Fiery, gaseous, aqueous and
earthy materials apparently originate from the forms in (b).
Matter in (d) takes the place of Aristotelian second substance. In the tradi-
tion of Boethius’ commentaries, genus is metaphorically considered the matter
of the species because it provides a kind of support for qualitative properties.
Just as form combined with matter gives rise to a body, e.g. a statue, similarly
specific difference added to genus constitutes species. According to this posi-
tion, however, second substances have a qualitative nature, and are an esse,
not an id quod. Subsistences, considered as matter in (d), are also forms in the
sense of (c). This fact causes some friction between the traditional interpreta-
tion and the present one.
The sense of the metaphor in the traditional interpretation is quite clear: an
object such as /animal/ is to rationality as marble is to the form of the statue.
‘Animal’ is a substantive, and we tend to think that it has a different meaning
from that of ‘rational’ (i.e. rationality). Gilbert believes only forms exist and /
animal/ is, or should be, the same subsistence as /animality/. Both rationality
and animality are subsistences; as we have seen, Gilbert confirms this position
in several different passages.
To consider a subsistence the matter of another can only mean that in the
predication tree a subsistence x comes before y, and that there exists a subsis-
tence z such that x + z = y. Seen as subsistences x, y and z appear to be objects
of the same type. On the other hand, Gilbert distinguishes between specific,
generic and differential subsistences. This difference is less visible than the
traditional one, which relies on a clear syntactic distinction between species/
genera and differences, but it may play some role, regardless of the connection
to the logical grammar. As we shall see in the next section, it is difficult to reach
a definition of what makes an object a subsistence, but the difficulty seems
more related to the difference between subsistences and accidental forms than
that between specific/generic and differential subsistences.
Another term Gilbert uses to refer to subsistences is ‘nature’. According
to Boethius’ Contra Eutichen et Nestorium, ‘nature’ is said with reference to
(a) bodily entities only, (b) bodily and incorporeal things, or (c) all things that,
being somehow conceived,24 are said to be. A further sense of ‘nature’ is (d) the
one by which silver is said to have a different nature from gold; in this case,
‘nature’ stands for the specific difference. Boethius provides definitions for
each of these different senses. Note that he considers the soul, God and divine
24 Boethius points out that the clause ‘somehow’ is used to include objects that are difficult
to conceive, such as God; ‘that are’ is used to exclude items that are not in any way, such
as what is referred to by the word ‘nothing’.
244 Chapter 9
things to be incorporeal natures; ‘nature’ in this case means ‘that which can act
or be acted upon.’
Gilbert gives an accurate report of Contra Eutichen up to point (c); his is
a literal commentary which does not stray far from the text. When the time
comes, however, to discuss (d), after the usual citation of Boethius’ words ‘est
etiam etc.’, he introduces some observations that do not seem to be a gloss, but
rather a deviation from the general plan of the commentary. The digression is
presented, or justified, as follows:
We must also consider, however, that this use of that noun [‘nature’] is
completely alien to the principles that do not proceed from an original
act of creation. The noun ‘nature’ has more to do with things that are
born, and is derived from ‘born’ (‘natus’), for the grammatical properties
of denomination (according to which a noun is taken from some other
expression, not without participation in the signified thing), even if the
same noun is not appropriate for all things that are born. It seems, in-
deed, that what is called ‘nature’ is the nature of something.25
Gilbert continues by arguing that not all things that are born can be called
‘natures’; rather, the noun is suitable for those which are natures of something.
The apparent petitio principii is explained as follows:
Things that are something – i.e. that subsist for something different from
themselves – are not natures of something, rather they are what they are
by reason of a nature. A body or a spirit are not [natures] of something,
but something is the nature of the body or the spirit. I believe the noun
‘nature’ is to be used for the subsistences of those [bodies and spirits]
rather than for the accidents of the subsistences. Although such acci-
dents are native and are of something [i.e., of the things to which they
adhere or are connected to in some way], since they are not beings of
subsistent things, and consequently are not able to determine generation
25 In Contra Euticen, 260, 80–85: ‘Sed attendendum quod a principiis, que nulla creationis
nativitate procedunt ab aliquo, huius nominis appellatio est omnino remota. Nativis
autem secundum grammatice denominationis proprietatem qua nomen ab aliqua dic-
tione non sine rei significate participatione assumitur, magis accedit et a natu ‘natura’
vocatur, quamvis non omnibus nativis hoc nomen recte convenire intelligatur. Videtur
enim id quod natura dicitur alicuius esse natura.’
Gilbert of Poitiers 245
and corruption by their own power, they are called ‘natures’ due to a non-
common use of a more subtle logic. 26
This diversion reveals another meaning of ‘nature’, which differs from the ones
considered by Boethius: natures in the proper sense are simply subsistences.
Gilbert adds that native things properly called ‘natures’ are the ones that con-
stitute the being of native subsistents, i.e. first created substances. After this
observation the commentary resumes from the point where it was interrupted
(the discussion of meaning (d) of ‘nature’), but the deviation is not without
consequences.
When Boethius speaks of corporeal and incorporeal substances, what he has
in mind are particular and universal substances (first and second substances).
Sense (d) of ‘nature’ – nature meant as specific difference – is, for Boethius,
really distinct from the senses previously considered, while for Gilbert both
second substances and differences are subsistences. This means that the de-
viation in the commentary is seemingly unavoidable: in fact, it is necessary to
provide a sense of nature that includes both second substances and specific
differences.
We conclude this part with an annotation about the occurrence of the ad-
jective ‘native’ in the last text quoted. This usage is particularly important be-
cause Gilbert’s position on universals is characterized by John of Salisbury as
one that emphasizes native forms. Gilbert does indeed sometimes speak of
the distinction between native and ‘genuine’ things, although this point cer-
tainly cannot be considered a leitmotiv of his position. Perhaps the most sig-
nificant text is a comment on the famous passage in De Trinitate on the degree
of speculative knowledge; Gilbert claims that native things are the proper ob-
ject of disciplinary or mathematical knowledge, while genuine things are the
subject of intellectual speculation. In two other short passages, he provides a
list of genuine things: in one he considers God and primordial matter, and in
26 In Contra Euticen, 260, 87–97: ‘Unde illa que vere sunt aliquid – hoc est que aliis quam
ipsa sint vere subsistunt – non sunt aliquorum nature, sed eorum potius aliqua sunt na-
ture. Non enim corpus aut spiritus aliquorum, sed magis corporis aut spiritus aliqua sunt
nature. Et puto quod hoc nomen eorum subsistentiis quam subsistentiarum accidenti-
bus accomodatius convenit. Quamvis enim accidentia et native sint et aliquorum sint –
eorum videlicet quibus insunt vel adsunt vel quolibet modo extrinsecus effiguntur – quia
tamen non sunt subsistentium esse et ideo illorum nec access generatio nec abscessu
corruption fieri potest, raro usu logice subtilioris vocantur ‘nature’.’
246 Chapter 9
the other God, matter, and prototypical ideas, which should precede – in some
metaphysical sense – native subsistences.27
9.3 Categorization
The ten and so on. Here it should be noted that, faculties differing depend-
ing on their different kinds of disciplines (e.g. of the natural sciences,
mathematics, theology, law, rational studies), that of the natural sciences
is more devoted to the study of natural language (in humane locutionis)
[…]. In natural sciences, ‘what quantity’ and ‘what quality’ are properly
said in the same sense as in sentences like ‘what quantity or quality a
man or stone has’ […]. In mathematics we say ‘what quantity or quality a
line has’, and in theology ‘what quantity or quality God has’, in law ‘what
quantity or quality a dignitary’s land has’, in moral philosophy we discuss
‘the quantity or quality of honesty’.28
27 There are not many texts available on ideas. Gilbert speaks of ideas that are inherent
in particular things and are specimens of what is found in them as images (99, 96). He
also makes some comments on the passage in which Boethius claims that forms in mat-
ter are called forms only in a secondary sense, while true forms are precisely the ideal
eternal specimens (100, 98). Finally, there is a text which seems to contrast the form-usia-
subsistence and its ideal specimen, a text which – if I interpret it correctly – leads straight
to an undesirable inflation of formal entities (cf. 195, 37).
28 In De Trinitate, 115, 1–12: ‘Decem etc. Hic commemorandum est quod, cum facultates se-
cundum genera rerum de quibus in ipsis agitur diverse sint, i.e. naturalis, mathematica,
theologica, civilis, rationalis, una tamen est, scilicet naturalis que in humane locutionis
usu promptior est […] ‘quantum’ et ‘quale’ in naturalibus recte dicuntur ut quantus et
qualis homo vel lapis. […] ut dicatur in mathematicis ‘quanta et qualis linea’ et in theo-
logicis ‘quantus et qualis Deus’, et in civilibus ‘quanta et qualis prepositura’, et in rationali-
bus ‘quanta et qualis honestas’.’
Gilbert of Poitiers 247
This noun ‘substance’ was not given on the basis of the natural genus,
but of the common notion of whatever is the being of subsistent things,
not just to beings, i.e. subsistences, but also to the things of which subsis-
tences themselves are beings, i.e. all subsisting things.30
Things belonging to the genus /substance/ are evidently not of the same kind
and/or logical level, since Gilber considers both individuals (subsisting things)
and their essential properties (subsistences). Gilbert says something more on
this sortal difference in the following text:
29 For example, ‘___est species’ occurs in a logical context, ‘Deus est___’ in a theological one.
On this topic see, inter alia, par. 7.4.
30 In De Trinitate, 116, 36–39: ‘Hoc nomen quod est ‘substancia’ non a genere naturalium sed
a communi ratione omnium que sunt esse subsistencium inditum est, non solum illis que
sunt i.e. subsistenciis, sed etiam illis quorum ipse sunt esse, i.e. omnibus subsistentibus.’
31 In De Trinitate, 117, 78–118, 91: ‘Quicquid enim est subsistencium esse, eorundem substancia
dicitur. Quod utique sunt omnium subsistencium speciales subsistencie et omnes ex qui-
bus he composite sunt […] simplices quidem alicui novem generum supponi, ut animatio
supponitu habitui, rationalitas qualitati. Compositas vero – ut est humanitas et cetere
speciales quas ex generalibus atque differentialibus constare nullus ignorant – dicimus
248 Chapter 9
I do not know who the logicians are who use ‘subsistence’ and ‘substance’
interchangeably. Gilbert certainly refers to second substances and proposes
a ‘mathematical’ interpretation of expressions like /man/, /animal/. The last
sentence of the above passage is difficult to interpret and translate; what ex-
actly does it mean that simple and composed things sunt eorumdem generum?
I have proposed a translation of ‘sunt’ as ‘consist’;32 the sense would therefore
be that, having rationality as a component, humanity consists of a quality. The
alternative – difficilior – is to interpret the sentence as if Gilbert were saying
that subsistences are of the same kinds as their components. Let us examine
the outcomes of both alternatives.
Whatever the case, it is clear that the species of substances and their qualita-
tive components are seen as subsistences. There is plenty of textual evidence
that confirms the disappearance or transformation of Aristotelian second
substances,34 and Boethius even provides some ideas for a ‘formal’ interpre-
tation of second substances. We recall here some observations encountered
along the way:
esse eorumdem generum quorum simplices ille sunt que ipsas componunt, ut humanitas
genere est quod animatio et item quod rationalitas.’
32 A Latin example of this usage is ‘classis mille et duecentarum navium fuit’ (C. Nepos).
33 The categorial interference is probably greater when the assignment is generalized. For
example in the case of geometric figures (the presumed genera being /essence/ (accord-
ing to Boethius’ De Arithmetica Institutione), /quality/, and /quantity/).
34 Commentaries, 93, 71; 135, 99; 258, 74–75; 261, 88; 262, 93–95; 269, 22–24; 280, 42. The only
counterexample of which I am aware appears in a text in which Gilbert says that the
genus is a collectio (312.118). We will consider this text in the paragraph on universals.
Gilbert of Poitiers 249
Logicians and grammarians may share some responsibility for the suppression
of second substances. In the tradition of the lectio nominis, authorities do not
speak of second substances but of appellative nouns, not of the species /man/
but of the common noun ‘man’ (and possibly of its meaning). The noun used
to indicate the meaning often coincides with that of the subsistence (‘man’ →
‘humanity’, ‘being man’).
Thus, the ontological landscape presented by Gilbert appears to be populat-
ed by individual subsistent things (coinciding with Aristotelian first substanc-
es) and a number of formal substantial and accidental elements. Consider the
following texts:
Things that are said to be in any sense are subsistents, subsistences, and
their accidents. [Boethius] – as is clear from what follows – also consid-
ered God and matter. […] It should be well understood that the subsistent
does not accord with subsistences or accidents for any genus or reason.
In fact, though the subsistent and subsistence are called substances or
35 I quote the whole passage (Rand, Theological Tractates 88, 42–55): ‘Nam quod Graeci
ousiosin vel ousiosthai dicunt, id nos subsistentiam vel subsistere appellamus; quod vero
illi ypostasin vel ypsistasthai, id nos substantiam vel substare interpretamur. Subsistit
enim quod ipsum accidentibus ut possit esse non indiget. Substat autem id quod aliis
accidentibus subiectum quoddam, ut esse valeant, subministrat; sub illis enim stat, dum
subiectum est accidentibus. Itaque genera et species subsistent tantum; neque enim acci-
dentia generibus et speciebus contingent. Individua vero non modo subsistent vero etiam
substant, nam neque ipsa indigent accidentibus ut sint; informata enim sunt iam pro-
priis et specificis differentiis, et accidentibus ut esse possint ministrant, dum sunt scilicet
subiecta.’
250 Chapter 9
subject things, they are such for different reasons. A fortiori, there exists
no common genus or reason that God or primordial matter share with
subsistent things and subsistences.36
[Boethius] presents the ten genera of predicaments. He then gathers
in two natural places, not all genera, but all predicaments; these places
are substance and accident. Every subsistent’s being is said to be its
substance. These things are the special subsistences of all subsistent
things, and what constitutes them, i.e. the general subsistences of their
subsistents, for which the [special] ones themselves are similar, and the
differential ones for which they are dissimilar.37
36 This text is a comment on what Boethius says about the concept of nature. In Contra
Euticen, 243, 20–30: ‘Res autem, que quocumque modo esse dicuntur, et subsistentes et
subsistentias et eis accidentes <sunt>; Deum quoque et ylen intelligit ut ex sequentibus
patet. […] Hic diligenter est attendendum quod subsistens cum subsistentia vel acci-
dentibus nullo prorsus genere seu ratione convenit. Nam etsi subsistens et subsistentia
dicuntur substantie et subiecte, alia tamen atque alia ratione. Molto magis ergo Dei et
primordialis materie cum subsistentibus et subsistentiis et accidentibus nulla est generis
aut rationis communio.’
37 In De Trinitate, 117, 76–83: ‘Predicamentorum omnium decem genera posuit. Nunc eadem
non dico genera sed predicamenta omnia in duobus locis naturalium colligit que sunt:
substantia et accidens. Quicquid enim est subsistencium esse, eorundem substan-
cia dicitur. Quod utique sunt omnium subsistencium speciales subsistencie et omnes ex
quibus he composite sunt, scilicet eorundem subsistencium per quas ipsa sibi conformia
sunt generales, et omnes per quas ipsa dissimilia sunt differentials.’
38 In De Trinitate, 90, 45–50: ‘De aliquo homine tota forma substancie qua ipse est perfectus
homo et omne genus omnisque differentia ex quibus est ipsa composita, ut corporalitas
et animatio et huiusmodi alie, et denique omnia que vel toti illi forme adsunt, ut humani-
tati risibilitas, vel aliquibus partibus eius, ut color, qui corporalitati, et scientia que adest
rationalitati et huiusmodi alia infinita.’
Gilbert of Poitiers 251
Quantities and qualities are distinct from other forms. Some adhere to the
overall form and to that of which it is composed (species, genera and differ-
ences). Others adhere to part of the total form and have a contingent nature,
such as color. Finally we have statuses, which extrinsically determine subsis-
tent things, i.e. individuals acting as a support to essential forms. Apparently
the categorization depends, or mainly depends, on the distinction between
essential and contingent properties and, within the latter, between ‘internal’
and ‘external’ contingent properties.
Gilbert’s notion of essential predication is not defined in detail; rather, it
seems to have to do with the natural order established at the time of creation,
an order subject to possible amendments according to divine resolution.40 On
the subject of generation and corruption, Gilbert observes that some proper-
ties are subsistences for species, but not for genera. In explication this point, he
considers the example of animation and incorporation:
39 In Contra Euticen, 319, 59 – 320, 71: ‘Natura enim subsistentis est qua ipsum subsistens aliq-
uid est. He vero sunt substantiales forme et que illis in ipso subsistente adsunt, qualitates
et intervallares misure. Cetera vero que de ipso naturaliter dicuntur, quidam eius ‘status’
vocantur, eo quod nunc sic nunc vero aliter, retinens has quibus aliquid est – mensuras et
qualitates et maxime subsistentias – statuatur. […] subsistentiis manentibus, homo nunc
hoc nunc illo situ vel loco vel habitu vel relatione vel tempore vel actione vel passione
statuitur, et, idem permanens, secundum extrinsecus sibi accidentia variatur.’
40 Gilbert spends several pages arguing that necessity is part of the history of creation; there
are things that always (i.e. in and outside of time) are such and such, while others are in
the course of time. Others, still, can be changed continually by divine intervention. Cf.
In Contra Euticen, p. 320 and following.
252 Chapter 9
and soul. Therefore, animation of the body and incorporation of the soul
appear to be subsistences. Indeed they are, but neither of the soul nor of
the body, but of that which is composed by them, i.e. the animal. In fact,
the soul perfectly subsists without incorporation, and the body is per-
fect without animation. The animal, on the other hand cannot be animal
without incorporation of the soul and animation of the body. For this rea-
son they (animation and incorporation of the soul) are extrinsic habits of
the body, while they are subsistences of the animal, of its generation and
corruption.41
Thus, subsistences are those properties which are essentially inherent in sub-
sistent things. In what way, however, and in how many things are they inher-
ent? What we would like to obtain is an intuition about what subsistences
actually are. Gilbert provides us with elements to determine what a form or a
property is in general. As for subsistences, though, we only have a character-
ization of the modal relation between subsistences themselves and subsistent
things. In other words, we cannot distinguish a substantial property from an
accidental one without considering what happens when they are predicated
of a subsistent thing. A way to do this may be to assume:
This definition does not appear to fully capture the concept of subsistence
which Gilbert has in mind. In principle, there may be properties that some-
thing has essentially while something other does not or has contingently.
Gilbert provides an example in the text just quoted: animation is an essential
property of the animal but not of the body, i.e. if a thing is an animal, then it
necessarily has a soul; if it is (only) a body, it has animation as an extrinsic
property. Thus, an inanimate body, and an animated one (an animal) would
meet the definition, i.e. both have animation, but the prediction that ‘anima-
tion’ is a subsistence is wrong because it is not always true. For an animal is,
41 In Contra Euticen, 320, 78–88 : ‘Cum enim corpus animatur vel anima incorporatur, fit hac
corporis et anime coniunctione generatio animalis; itemque corporis et anime disiunc-
tione eiusdem animalis corruptio. Unde animatio corporis et incorporatio anime sbsis-
tentie esse videntur. Et sunt utique, sed neque anime neque corporis sed illius quod ex his
compositum est animalis. Anima namque preter sui incorporationem perfecte est anima.
Et corpus preter animationem perfecte est corpus. Animal vero nec est nec potest esse
animal preter anime incorporationem et corporis animationem. Ideoque anime et corpo-
ris sunt extrinseci habitus, animalis autem sue generationis et corruptionis subsistentie.’
Gilbert of Poitiers 253
in fact, a subsistence, while for a body without a soul it is, in some sense, an
extrinsic property.
Another example of a property that something has essentially and some-
thing else has contingently is mortality. Man will be immortal after resurrec-
tion; therefore, mortality is for man a transient property (Gilbert would say
a status), while it can be a subsistence for living things that do not share the
same project God established for man.
One can try to use a weak definition of subsistence:
9.4 Particulars
When Gilbert talks about subsistences, the terms he uses most often (e.g. ‘hu-
manity’ or ‘rationality’) must be understood as proper nouns. However, not
everything goes in this direction. For example, Gilbert says that in the process
of mathematical abstraction forms are obtained ‘from both that in which they
are and themselves’,42 suggesting that their non-abstract counterparts actually
are particulars.43 Some interpreters believe that these particular properties
play a role in Gilbert’s ‘solution’ to the problem of universals.44 It must be said,
though, that John of Salisbury, who first expressed this idea, does not appear
to be entirely certain, if I interpret correctly what he says about the fact that
Gilbert (or someone close to him) asserts the conformity of native forms.45 I
do not believe that the texts in which Gilbert talks about individual proper-
ties can belie the ones where generic and specific subsistences are considered.
One must fully understand, however, what relationship there exists between
particular and general subsistences, and consequently, how Gilbert’s position
on the problem of universals can be determined.
Of the texts in which Gilbert explicitly refers to individual properties and
subsistences, let me quote the following:
Poitiers e nella scuola porretana,’ Documenti e studi sulla tradizione filosofica medievale, 19
(2008), 191–246; 200–215. C. Erismann, ‘Explaning exat resemblance. Gilbert of Poitiers’s
conformitas theory reconsidered’ in R. Pasnau (ed.), Oxford Studies in Medieval Philosophy,
vol 2, OUP, Oxford 2014, 1–24. Some of the scholars mentioned attach importance to some
critical observations by Theodoric of Chartres on the plurality of forms. I do not believe
that Theodoric’s texts are unequivocal: Gilbert is not the only one who speaks of indi-
vidual essences, so do Joscelin, supporters of TR2 and TR3 (see the chapter on realism).
Even if it were true that the target of Theodoric polemic is Gilbert, as the aforementioned
authors argue, it is questionable whether Gilbert’s overall theoretical position is at issue.
45 John of Salisbury, Metalogicon, ed. J.B. Hall, Brepols, Turnhout 1991, 83, 84–86 (PL 199, 875
D): ‘Porro alius, ut Aristotelem exprimat, cum Gilberto episcopo Pictavensi, universalita-
tem formis nativis attribuit, et in earum conformitate laborat.’
46 In De Trinitate, 58, 42 – 59, 52: ‘Est enim proprium naturalium quod ‘sicut numero diver-
sorum proprietates diverse sunt ita quoque subsistentie numero sunt diverse’ et quod
‘una singularis subsistencia non nisi unum numero faciat subsistentem’, ut Platonis et
Gilbert of Poitiers 255
The ‘bleeding’ metaphor concluding the quotation does not entirely hide the
fact that there is a discontinuity between the subsistence of the body (that
is, corporeality) and the single subsistence ‘incarnated’ in the generating and
generated bodies.
One of the points that seem to be particularly significant for the comparison
between universal and particular is the comment on the text in which Boethius
speaks of identity and similarity. Gilbert maintains the following thesis:
Cato is the same as Cicero, since naturally the species of Cato and Cicero,
that is /man/, is the same. Several subsistences, which are one species,
one of which is man-Cicero, the other man-Cato, make them substan-
tially similar.47
The being is always different, at least in number, and its accidental
property pertains to different things, as every human being is different by
number only and not by dissimilarity. Some accidents are different by
number only – such as whiteness and the whiteness of two white things –
some even by dissimilarity, such as whiteness and the blackness of a
black and white thing. The manifold being of men or horses and its acci-
dents are not only different in number, but are also dissimilar.48
Ciceronis non solum accidentales proprietates, verum etiam substantiales, quibus ipsi
sunt verbi gratia vel diversa corpora vel diversi homines, diverse sunt. Et quecumque sin-
gularis proprietas Platonem corpus esse vel hominem, eadem nullum alium idem esse
facit. Nam etsi generationis propagatio generato dat esse corpus ab ea singulari corporali-
tate transfusi sanguinis, a qua generans corpus fuerat, fit tamen decisione discessio qua
et generans ab illius corporis, quod decisio sanguine retinet, subsistentia et generatus ab
illius, quod deciso sanguini fomentis temporalibus additur, aliud ab invicem corpus fit.’
The text presents some difficulties, as is demonstrated by the many variations in critical
apparatus.
47 In De Trinitate, 75, 32–35: ‘Idem est Cato quod Cicero, quia scilicet et Catonis et Ciceronis
eadem species est verbi gratia, ut homo. Diverse namque subsistentie, que una sunt spe-
cies, quarum alia Cato alia Cicero homo est, eosdem substantialiter faciunt similes.’
48 In De Trinitate, 100, 41 – 101, 47: ‘Nam diversum – saltem numero vel etiam dissimilitu-
dine – esse et ei accidens semper est diversorum: ut hominum non dissimilitudine, sed
solo numero diversum est omne esse. Accidentia vero quedam numero solo – ut duorum
alborum albedo et albedo – quedam etiam dissimilitudine – ut albi et nigri albedo et
nigredo – diversa sunt. Hominum vero atque equorum plurimum esse et ei accidens non
solo numero sed etiam dissimilitudine diversum est.’
256 Chapter 9
being man (in) Cato and being man (in) Cicero, which, both being human sub-
sistences, are similar. Moreover Cato and Cicero are different by number and
dissimilarity. In the same way, being man and being horse are dissimilar but the
being animal of man and the being animal of horse are not dissimilar. The dif-
ferent cases are illustrated in the following table:
I will use notation like ‘thisi being man [being man, Mi]’ to denote a particu-
lar subsistence defined by its generic (being man) and individual components
(Mi). Consequently, a general subsistence will be indicated by a script like
‘[being man]’. The fact that two individual subsistences are numerically differ-
ent does not prevent their ‘bearers’ from being identical in species, as stated by
Boethius and confirmed by Gilbert. One may wonder whether the identity of
the species of Plato and Cicero involves, on one side, the identity of the genus
of Plato and Brunellus, on the other of thisi being man and thisj being horse. It
does, however seem difficult to believe that Gilbert accepts the idea that dis-
similar things are identical, albeit sharing some subsistences.
Further information on individual subsistences and their similarity can
be obtained from the following text (a digression from the Boethian topic of
‘person’):
49 In De Trinitate, 143, 53 – 144, 60: ‘Alicuius proprietas alia ratione singularis, alia individua,
alia personalis vocatur […] In naturalibus enim quicquid est alio quam ipsum sit, aliquid
est. Et quoniam id quo est aliquid singulare est, id quoque quod eo est aliquid singu-
lare est.’
Gilbert of Poitiers 257
This definition of singularity leaves open the possibility that abstract entities
can be considered singular, i.e. properties can be predicated of other proper-
ties. ‘Dividual’ things are characterized as follows:
Gilbert speaks of ‘dividual’ things, and this would lead to the exclusion of indi-
viduals, meant as first substances. Subsistences, however, may present them-
selves either in a super-singular form, so to speak, or a singular one: we can
speak of the humanity of Socrates or of humanity. According to Gilbert, di-
vidual things are ‘not only those things that are, but also those to which they
conform.’ Evidently the things that are cannot be subsistent things, namely
first substances, inasmuch as these are individuals; ‘things that are’ can only
mean instantiated subsistences (like the being-man of Socrates); whereas
type-subsistences (like being-man) are those by reason of which the instanti-
ated things are similar.
A consideration follows which sheds some light on the metaphysical struc-
ture of singular entities:
It must be considered that those things for which that which is, is some-
thing, are either simple like rationality, or composed like humanity.
Simple things either in act or by nature are of the same form, therefore
50 In De Trinitate, 144, 63–68: ‘Sepe … diversa numero singularia secundum aliqua eorum
quibus sunt conformia sunt. Ideoque non modo illa que sunt, verum etiam illa quibus
conformia sunt unum dividuum sunt. Ac per hoc neutrum illorum quibus conformia sunt
illa que sunt individuum est. Si enim dividuum facit similitudo, consequens est ut indivi-
duum faciat dissimilitudo.’
258 Chapter 9
– Conformed things are simple or composed, from some types of things but
not all.
– Individuals do not conform, or more precisely, do not conform to anything
other than themselves.
– Individuals are the ‘product’ of things of all types.
51 In De Trinitate, 144, 69–78: ‘Attendendum vero quod ea quibus id quod est est aliquid, aut
simplicia sunt ut rationalitas aut composita ut humanitas. Simplicia omnia vel actu vel
natura conformia sunt. Ideoque nulla eorum vera dissimilitudinis ratione sunt individua.
Composita vero alia ex aliquibus tantum, alia ex omnibus. Que non ex omnibus similiter
sicut et simplicia vel actu vel natura conformia sunt. Ac per hoc nulla eorum sunt indi-
vidua. Restat igitur ut illa tantum sint individua que ex omnibus composita nullis aliis
in toto possunt esse conformia: ut ex omnibus que et actu et natura fuerunt vel sunt vel
futura sunt Platonis collecta platonitas.’
Gilbert of Poitiers 259
The interpretation of these data is not unique. What we propose below consid-
ers subsisting things and subsistences as having a ‘bundle’ of subsistences; an
entity conforms to another entity if it ‘bundles up’ all the elements it owns.
Here a linguistic problem arises: the Latin ‘esse conformis’ corresponds to the
English ‘to conform-to’. The intuitive sense of the conformity relation is bot-
tom-up, i.e. from the tokens to the type. However, in Gilbert’s technical lan-
guage, dividual things are conformia, thus the relation should be reversed from
types to tokens. Hence I would suggest using ‘to conform-with’ to indicate the
top-down relationship, and ‘to conform-to’ for the inverse relationship. I also
think it is appropriate to distinguish between broad and narrow conformity.
Indeed, Gilbert claims individuals cannot conform in their entirety with any-
thing else (Nullis aliis in toto possunt esse conformia), leaving the door open for
self-conformity.
On the one hand ‘conformity’ seems to indicate a relation; on the other, a
property, since the second term of the relation is missing (as in the text above).
In the following definitions, Fx stands for the set of subsistences of x:
X conforms-to Y iff Fy ⊆ Fx
X strictly conforms-to Y iff Fy ⊂ Fx
Below are some examples of beings, together with their metaphysical compo-
nents (entities and their related sets of properties are co-indexed; strict confor-
mity implies simple conformity):
(cont.)
being X
Conforms to Conforms to
All that is, is singular, but not everything which is, is individual. Among
singular things some are similar to others because of their total property,
and because of this conformity, all these at the same time are called di-
vidual, just like different qualities of the different bodies are called qualia
by reason of their entire species.52
52 In Contra Euticen, 270, 73–77: ‘Quicquid enim est singulare est. Sed non quicquid est in-
dividuum est. Singularium namque alia aliis sunt tota proprietate sua inter se similia,
Gilbert of Poitiers 261
It does not seem here that dividual things are particular qualities/properties,
but rather some plural conjunction of them. The problem is that there is no in-
dication that Gilbert accepts a collection theory in the style of Joscelin. For this
author universals are collective totalities formed by individual essences. I do
not think, therefore, that it is appropriate to attribute a rather elaborate theory
to Gilbert only because in some circumstance he appears to be speaking of
collective totalities. I would rather give a ‘weak’ – and ontologically cheaper –
reading of the phrase ‘all these at the same time’ above, considering the ‘total
property’ as that (subsistence) which allows for gathering ‘the different (par-
ticular) qualities of the different bodies’. All things under a species would not
be a collectio or dividuum but be so called (dicuntur unum dividuum) because
their similarity.
From what Boethius says later, it seems clear that the forms subject to math-
ematical knowledge are also those corresponding to the genera and species of
que simul omnia conformitate huius ratione dicuntur unum dividuum: ut diversorum
corporum diverse qualitates tota sui specie quales.’
53 I quote from Häring’s edition, based on the text of the Commentaries (cit.), 371, 3–12: ‘Cum
tres sint speculative partes: naturalis, in motu inabstracta anypezairetos, i.e. inseparabilis.
Considerat enim corporum formas cum materia que a corporibus separari non possunt.
[…] Mathematica sine motu inabstracta. Hec enim formas corporum speculatur sine ma-
teria ac per hoc sine motu. Que forme cum in materia sint, ab his separari non possunt.
Theologica sine motu abstracta atque separabilis.’
262 Chapter 9
the substance (like /humanity/ and /animality/). Moreover, from Gilbert’ com-
ments we know that there is a natural interpretation of the species and genera
mentioned by Porphyry in terms of ‘mathematical’ forms.
The commentary on De Trinitate does not closely follow the text, and con-
tains several digressions. One important point is the distinction between the
different meanings of form and matter. As we have seen, subsistences are con-
sidered in one sense forms, and in another matter of subsistent things. Yet
forms are precisely the subject of intellectual abstraction, as confirmed by the
following passage:
The forms at issue are subsistences and other qualitative features. While em-
phasising that these forms do not exist in nature, Gilbert nevertheless makes
them the main object of his philosophy (or his Boethian hermeneutics). There
is a point in the text that could lead to a proliferation of abstract entities: so
far has appeared that the name of a subsistence is ‘being x’. There are subsist-
ing things and subsistences for which subsisting things are what they are; for
example, there are men and there is the subsistence being man, by reason of
which men are men. On the other hand, subsistences themselves are present-
ed as a kind of singular objects, with some form of objectivity; as such, one may
ask what determines their being such and such (and being known as such and
54 In De Trinitate, 84, 70 – 85, 81: ‘Alia vero speculatio que nativorum inabstracta formas aliter
quam sint, i.e. abstractim considerat, ex fine quo illud facit Grece quidem mathematica,
Latine vero disciplinalis vocatur. […] Neque enim rationalis speculatio perfecte id quod
est esse aliquid capit nisi disciplinalis quoque id unde illud est quid sit firmiter teneat.
Verbi gratia, non perpendit ratio quid sit esse corpus et esse coloratum te esse latum nisi
disciplina quid sit corporalitas quid color quid latitudo cognoscat. Quod fieri non po-
test nisi hec inabstracta atque concreta et ab eo in quo sunt et a se invicem abstrahat et
discernat.’
Gilbert of Poitiers 263
such). Gilbert claims that this point can be interpreted in different ways: a) for
each subsistence, an abstract form different from it must be imagined; b) any
subsistence, as an object, falls under a higher subsistence. As a post-Ockhamist
and post-Fregean interpreter, I would be inclined to choose path (b). Of course,
the price to pay for the latter choice is that the meaning of ‘fall under’ must be
left relatively vague.
At this point we have collected a sufficient number of data to articulate the
Porphyrian problem. For Gilbert there is only one thing – subsistence – that
could play the role of species or genus. The concept of subsistence, however,
is underdetermined: at the most we can say that subsistences – since they do
not exist in the same way as particular things – are the subject of intellectual/
mathematical abstraction. In order to take a step further we must attend to the
only passage in which the author speaks of universals; unfortunately the texts
quoted below are not easy to grasp (nor to translate).
Beginning with the second text, we notice that Gilbert speaks precisely of spe-
cific generic subsistences that are, in some sense, plurals. The first text does
not speak explicitly of subsistences, but it is quite clear that these are identical
to second substances. Strikingly, in both of these texts subsistences are called
by name, and it is said that they are qualities of nouns. This characterization
is standard for dialecticians of the twelfth century; it originates from Priscian’s
definition of noun as that which means substance and quality. For example,
55 In Contra Euticen, 269, 35–50: ‘Substantiarum alie sunt universales substantialis
forme similitudine, alie sunt particulares i.e. individue plenarum proprietatum dis-
similitudine. […] Universales sunt que plures secundum se totas inter se suis effecti-
bus similes de pluribus singulis subsistentibus inter se vere similibus praedicantur.
[…] Nam et homo videlicet subsistentia specialis, que est huius nominis qualitas, una
quidem conformitate sed plures essentie singularitate, de singulis hominibus, et
animal de singulis animalibus […] dicuntur.’
264 Chapter 9
‘man’ is said to mean men as substance and being man as quality. So far so
good; however, the same universal substances are also plures, i.e. not discrete.
A ‘strong’ interpretation of these texts runs parallel to the one proposed in
the previous paragraph: subsistences are to be considered as those pluralities
we had discussed. Here, too, for the same reasons already mentioned, I prefer
a ‘weak’ interpretation: a subsistence is ‘plural’, because either it has several in-
stances (or ‘tokens’) or it is predicated of several individual things. ‘Plural’, just
like ‘universal’, is a characterization which does not refer to what a subsistence
is or is not in itself, but to its being in relation to other things. I am by no means
entirely comfortable with this interpretation, and I still have particular doubts
as to Gilbert’s competence to discuss universals at a time when every word was
tested by dialecticians so as to be used against the proposer.
I have talked about several different perspectives or approaches to
Porphyry’s problem; I will now examine that of predication frequently adopted
by dialecticians contemporary with Gilbert. The first text that I would like to
mention closely follows the ones on universals. Interestingly, without warning
the reader, Gilbert moves from the linguistic level (omnipresent in all his com-
mentaries) to the metalinguistic level:
Dialecticians call ‘man’ and ‘sun’ dividual, grammarians call them appel-
latives. […] ‘Man’ by nature and in act is dividual and appellative; ‘sun’ by
nature only and not in act. Indeed, many things not only by nature but in
act were, are and will be men by a substantial similarity. Many things do
not happen to be suns in act but still are suns by nature […]. ‘Man’, from
some subsisting men in act and by nature, and ‘sun’, from things not in
act but by nature, are nouns of things similar to each other for the whole
substance of the form.56
56 In Contra Euticen, 273, 50–62: ‘‘homo’ et ‘sol’ a gramaticis appellativa nomina, a dialecti-
cis vero dividua vocantur. […] ‘Homo’ tam natura quam actu appellativum vel dividuum
est; sol vero natura tantum non actu. Multi namque non modo natura verum etiam actu
et fuerunt et sunt et futuri sunt substantiali similitudine similiter homini; multi quoque
numquam actu sed semper natura similiter soles. […] ‘Homo’ quidem ab aliquibus homi-
num subsistentis tam actu quam natura, ‘sol’ vero ab aliquibus non actu sed sola natura,
inter se invicem tota substantia forme similibus nomina sunt.’
Gilbert of Poitiers 265
changed: the noun applies to a number of things that were, are and will be
such and such, in accordance with the logical-grammatical tradition.
As for Priscian’s maxim ‘every noun signifies quality and substance’, a par-
ticularly interesting additional passage is worth mentioning in view of the sub-
sequent discussion (late 12th–early 13th century) on the contextual meaning of
expressions:
Each noun [or adjective] means different things, i.e. substance and qual-
ity, like ‘white’, means that which is called white – that is, the substance
of the noun – and that by which (id quo) it is called white – that is, the
quality of the same noun. If someone says, ‘a white [thing] is a body’, or
‘white is an accident’, whoever hears this same noun ‘white’, considering
the properties of things, will interpret the former sentence as if it were
about that which is called white, which actually is a body, the latter one
as if it were about that by which it is called white, which actually is an
accident.57
Incidentally, this is one of the few places where an accidental quality is called
id quo (see above for an analysis of this aspect). The fact that in this case the
quality is accidental is to be considered an accident. Once again, we would like
to draw attention to the fact that the meaning of a term in suppositio simplex,
(to anticipate a later terminology) is the quality, not just a particular instance
or a cluster of properties.
A large part of the dialecticians’ intellectual energy is spent explaining the
meaning of the relationship between subject and predicate of a categorical
sentence, with the aim of determining its truth or falsity. The explanations
given clearly presuppose a certain attitude about the problem of universals.
Here Gilbert has no theory to present, not even in the form of digression.
However, we are able to produce some textual evidence consistent with an in-
terpretation of predication in terms of inherence:
Another use of the noun ‘matter’ is made, with reference to general and
special susbsistences that are said to be beings of the subsisting things in
57 In Contra Euticen, 297, 57–64 : ‘Omne vero nomen diversa significat, substantiam videlicet
et qualitatem, ut ‘album’ id quod appellatur album – quod est substantia nominis – et id
quo appellatur album – qud est eiusdem nomins qualitas. Sub eodem igitur nomine quod
est ‘album’ si quis dicat ‘album est corpus’, ‘album est accidens’, rerum proprietatem secu-
tus interpres illud, sicut dictum est, de eo quod appellatur album quod vere est corpus,
hoc vero de eo quo appellatur album, quod vere est accidens, interpretabitur.’
266 Chapter 9
which they are […]. Corporality, which is the being of the body in which
it is, and for which the same is, of course, something, i.e. a body […], is
called ‘matter’.58
When [pure substances] are said to inhere in the objects of percep-
tion, it is not because they inhere in and adhere to them as non-abstract
entities, like corporeality inheres in the body.59
When it is said that something is predicated, one should not under-
stand that it is predicated of what is a species, but rather of the subsisting
thing in which the special subsistence is.60
Accidents and substances, both subsisting things in which accidents
are, and subsistences to which accidents adhere, can all be grasped by the
intellect.61
Notice that he said ‘essences’ above, he now says that substances are in
particular things […]. Corporality in act is in the body only, and the body
is not what it is said to be if corporality, which is its being, is not found
in it.62
In texts like these, subsistences are called by name. In some cases Gilbert
points out they are either special or general; he also says that the same enti-
ties are in subsisting things. We do not have a theory of inherence, but only
formulations that presuppose a certain hidden or pre-theoretical acceptance
of the concept of inherence. This concept certainly must be articulated, and
there are several ways to do so. The approach of classical realism is to imagine a
relationship between universals and particular entities, such that the universal
is located within the particular. Surely this cannot be Gilbert’s view, if not for
other reasons, because he devoted a number of pages to discussing particular
58 In Contra Euticen, 81, 73–84: ‘Ad generales quoque et speciales subsistentias que subsisten-
tium in quibus sunt esse dicuntur eo quod eis ut sint aliquid conferunt, eiusdem nominis,
i.e. materie, alia fit denominatio. […] corporalitas que est esse corporis in quo est et qua
ipsum est aliquid naturaliter, i.e. corpus […] ‘materia’ dicitur.’
59 In De Trinitate, 82, 22–24: ‘(Sincere substantie) sensibilibus inesse dicuntur, non ideo est
quod illis insint atque hereant inabstracte, qualiter corporalitas inest corpori.’
60 In De Trinitate, 93, 32–34: ‘Non enim […] intelligendum est de eo quod species est aliquid
predicari sed de subsistente pocius in quo est subsistentia specialis.’
61 In Contra Euticen, 245, 68–70: ‘Accidentia et substantie, tam subsistentes in quibus ac-
cidentia sunt, quam subsistentie, quibus adsunt, intellectu capi possunt.’
62 In Contra Euticen, 278, 7 – 279, 12: ‘Attende quod, cum superius dixerit ‘essentias’, nunc
dicit substantias in particularibus esse. […] Actu namque corporalitas nichil est nisi sit in
corpore, et corpus non est quod vocatur nisi in ipso sit corporalitas que est eius esse.’
Gilbert of Poitiers 267
subsistences and their relationships. There are other ways, however, some of
which are compatible with a nominalist or quasi-nominalist approach to the
problem of universals.
An approach to inherence is to consider the relationship between the en-
tity denoted by the grammatical subject, and singular entities like collections
or individual subsistences. Gilbert speaks of this relationship in the following
text:
the father is god, the son is god, the holy spirit is god. […]
then the father, the son and the holy spirit are one God by
the singularity of the essence, not three gods […] Conversely, it hap-
pens that Plato is a man, Cicero is a man, Aristotle is a man, and Plato,
Cicero and Aristotle are three men, not one by the singularity of the sub-
sistence. […] Not only is the second assertion about another person, but
it actually asserts something different, and the third one something still
more different in relation to the first and second ones. Although in fact
the second and the third assertions repeat the same predicative expres-
sion as the first, they do not repeat the thing predicated. The sentences
state different natures of things numerically different between them
(though they conform, or rather are numerically different, insofar as they
conform to something). […] It is not allowable to say that Plato, Aristotle
and Cicero are singularly one man.63
A possible reading of this text is that sentences like ‘Plato is a man’ and ‘Cicero
is a man’ express the propositions ‘Plato has this human subsistence’ and
‘Cicero has this other human subsistence’. For some interpreters, this formula-
tion is congruent with Gilbert’s concern for universal meanings.64 There are,
however, many aspects of this approach that simply do not add up: first, the
63 In De Trinitate, 72, 53–65: ‘Pater est Deus, Filius est Deus, Spiritus Sanctus est
Deus. […] Igitur Pater et Filius et Spiritus Sanctus sunt unus singularitate
essencie Deus, non tres Dii, […] Contra evenit ut Plato est homo, Cicero est homo,
Aristotiles est homo, igitur Plato et Cicero et Aristotiles sunt tres homines non unus sin-
gularitate subsistencie homo. […] non solum de alio, sed et singularitate sui aliud dicitur
secunda affirmation quam prima et aliud tercia quam prima vel secunda. Quamvis enim
secunda et tercia prime predicativum reppetant nomen, rem tamen predicatam non rep-
petunt. Sed quamvis conformes tamen diversas – immo quia conformes ergo numero
diversas – a se invicem naturas de [numero a se] diversis affirmant […] Non patitur hanc
adunationem ut dicatur: ‘Plato et Cicero et Aristoteles sunt unus singulariter homo.’
64 Among others, C. Erismann, op. cit..
268 Chapter 9
texts we have mentioned on predication do not say such things; indeed, among
Gilbert’s contemporaries, even philosophers who do not propose abstract en-
tities do not believe that the meaning of a predicate changes with each ut-
terance. The only author who might argue something similar is Joscelin, who,
however, discusses the issue within a rather complex theory which does not
exclude shared timeless meanings.
The particular context into which the quoted text is inserted needs to be
considered. As is clear from the last sentence, Gilbert is trying to support the
thesis that the things of which ‘man’ is predicated do not necessarily coincide.
However, the same cannot be said in theology. When we say that the three
Persons of the Trinity are God, we do not intend to claim that they are three
Gods, but one only. In the case considered, Plato, Cicero and Aristotle ‘are not
one by the singularity of the subsistence’. The unpleasant situation in which
the subjects are identical is avoidable if one assumes that the copula functions
as a sign of identity in utterances concerning God only (cf. Boethius) or, alter-
natively, that singular statements which are not about God are true if the dif-
ferent things ‘subject to’ the predicate expression (= res predicatae) share the
same property or nature.
In conclusion, I doubt that the text quoted provides evidence that the
author embraces an (extreme) ontological particularism. However, the pas-
sages where Gilbert admits of specific/generic subsistences (when he speaks
of metaphysics) as well as the marginal gloss on Priscian should not be over-
estimated either. We have seen that there are some important texts in which
Gilbert insists particularly on the maxim ‘everything that exists is singular’ and
proposes an ontological catalogue which includes both substances and indi-
vidual properties. What he does not do is provide a theory that explains exact-
ly which entity one must rely upon in speaking of truth conditions. I suspect
that introducing particular properties while still talking of generic and specific
subsistences is a symptom of the time in which Gilbert lives: rather than risk
proposing a losing theory, it is better to keep it in the rather vague and smoky
terms of the Boethian formulation, and at the same time suggesting that what
really exists is definitely individual.
I shall conclude with a summary of Gilbert’s ontology, as well as a sugges-
tion. His ontological catalog is quite varied: there are special, general and dif-
ferential subsistences, which are called by name: e.g. ‘humanity’, ‘animality’
(or ‘being man’ ‘being animal’), ‘rationality’. We then have all the qualitative
and quantitative determinations that rely on subsistences; these properties
have particular instantiations, yet at the same time eternal specimens (see the
reference to ideas or ‘pure substances’). In parallel, Gilbert describes a world
Gilbert of Poitiers 269
10.1 Background
In his main work, the Metalogicon, John of Salisbury1 devotes some space to
the issue of universals; he considers the various views and expresses his opin-
ion on Porphyry’s problem. The Metalogicon2 represents a kind of boundary
marker: the setting John provides serves the dual purpose of reconsidering the
issues faced by different authors (adding some historical and sociological de-
tails), and providing a glimpse of the future developments of the discussion.
In fact, John summarizes 50 years of interpretations of Porphyrian text while
at the same time providing some ideas (though in a rather convoluted way) for
new research perspectives.
Let us start from Book Two, Chapter 10 of the Metalogicon, in which the
author relates his own cultural education.3 John reaches Abelard’s school as
a student, or more precisely, as one ‘traveling for the sake of study’, and ‘at his
feet’ he learns the first elements of dialectic. John then follows the courses of
Alberic of Reims and Robert of Melun. It seems clear through his presentation
that John is not, ultimately, very attached to Abelard. His report reads like a
1 General information about John can be found in K. Jacobi, ‘Logic (ii): The Later Twelfth
Century’, in P. Dronke (ed.), A History of Twelfth-Century Western Philosophy, Cambridge
UP, Cambridge 1988, pp. 227–235; C.H. Kneepkens, ‘John of Salisbury’, in J.E. Gracia –
T.B. Noone (edd.), A Companion to Philosophy in the Middle Ages, Blackwell, Oxford 2003,
392–396; K. Guilfoy, ‘John of Salisbury’, Stanford Encyclopedia of Philosophy, 2005; F. Lejeune,
Introduzione a: Metalogicon. Jean de Salisbury, Presses de L’Université Laval, 2009, 2–101.
C. Grellard – F. Lchaud (ed.), A Companion to John of Salisbury, Brill, Leiden 2014; A selection
of studies on different aspects of John’s work and his cultural environment is contained in
M. Wilks (ed.), The World of John of Salisbury, Blackwell, Oxford 1984. On John’s place within
the Aristotelian tradition, see Ch. Burnett, ‘John of Salisbury and Aristotle’, Didascalica 2
(1996), 19–32, available at: www.scribd.com/doc/26824811/John-of-Salisbury-and-Aristotle-
Charles-Burnett-Didascalica-2-1996. A bibliography is available at the web address: http://
policraticus.hypotheses.org/files/2012/06/Bibliographie-Jean-de-Salisbury.doc.
2 Ioannis Saresberiensis, Metalogicon, ed. J.B. Hall, Brepols (Corpus Christianorum. Continuatio
Mediaevalis XCVIII) Turnhout 1991.
3 For a dating of the various stages of John’s education cf. K.S.B. Keats-Roha, ‘The Chronology
of John of Salisbury’s Studies in France: a reading of ‘Metalogicon’ II.10’, Studi medievali, 28
(1087), 193–203.
4 By the way Alberic who ‘inter ceteros opinatissimus dialecticus enitebat … revera nominalis sec-
tae acerrimus impugnator’, is a sworn enemy of Abelard’s. On Alberic’s cultural environment,
see J.R. Williams, ‘The Cathedral School of Reims in the Time of Master Alberic, 1118–1136’,
Traditio 1956 (vol. 12), 87–148.
5 Cf. L. Minio-Paluello, Twelfth Century Logic. Texts and Studies 1. Adam Balsamiensis
Parvipontani, Ars disserendi (Dialectica Alexandri), Edizioni di storia e letteratura, Roma 1956.
6 According to some scholars, William assumes the principle ‘ex contradictione quodlibet sequi-
tur’ (ECQ), and John, in refusing it, puts himself in a paraconsistent perspective. Kneale inter-
prets the reference to the ‘machine’ as a metaphor to describe the strength of an argument
(Kneale, W.C. – Kneale, M. (1962). The Development of Logic, Clarendon Press, Oxford, 201).
M. Amini inserts Kneale’s proposal into his short history of logical devices, suggesting that the
‘machine’ John talks about could have been a physical computing system (‘Logical Machines:
Peirce on Psychologism’, Disputatio, II-24 (2008), 335–348). On ECQ and its non-acceptance,
see, among numerous others: G. Priest – K. Tanaka, ‘A Brief History of ex contradictione quod-
libet’: Cap. 3 di: ‘Paraconsistent Logic’, Stanford Encyclopedia of Philosophy, 2009; G. Priest,
‘Paraconsistent Logic’, in D.M. Gabbay – F. Guenthner (eds.), Handbook of Philosophical
Logic, Vol. 6, Kluwer 2002, 287–393 (Cf. in particular pp. 293–294); id., ‘Paraconsistency and
John of Salisbury on Universals 273
After some experience teaching, John goes back to studying logic and theol-
ogy with Gilbert of Poitiers for a short time. He does not seem to have been
much impressed by his next masters, Robert Pullen and Simon Pexiacensis,
who both taught him theology. The lectures of the former amounted to little
more than life tips, and the latter was a faithful reader of the texts but was
‘weak in argumentation’. Thus ends the story of John’s cultural education. The
chapters immediately following this account of John’s twelve years of study
are a kind of general introduction to dialectics. His summary of the different
views on universals is contained in Chapter 17, entitled ‘What is taught in a
malicious way, and what are the opinions of contemporaries on species and
genera’. Even before introducing the various theoretical positions on Porphyry,
John issues a sort of generalized condemnation of the masters who, as a form
of conceitedness, present their positions in a convolute way without providing
useful learning tools.
This judgement concludes the historical-expository part of the section. The
teachers (John makes no further distinctions between them) do not care about
the learning difficulties and – more seriously – misunderstand Aristotle’s
thought, presenting it as closer to Platonism than it actually is or proposing
wrong interpretations. The reader receives an impression of John as a student,
and then a mature scholar, who is divided between his interest in what his
famous masters taught, and a certain distrust of their excesses both in life and
in doctrine. Perhaps as a result, John’s approach to the problem of universals
comes across as disinterested in or distrusting of traditional approaches.
Having followed this same path, albeit somewhat delayed and as a non-
attending student, I wish to add some of my own historical and sociological
considerations to John’s. I have nothing to say about his complaint of a lack of
bridging courses, except perhaps to note that the problems in higher educa-
tion that began at the time of Metalogicon have remained largely unchanged
over the course of the subsequent nine centuries. As for the contenders’ at-
tention-seeking, or, as John more benignly puts it, the fact that there were as
many men as opinions, the impression given is that in the shared setting of the
Porphyrian problem, almost all possible solutions have been proposed. Having
noticed, however, that every teacher wants to propose his own solution, we
Dialetheism’, in D.M. Gabbay – J. Woods (edd.), Handbook of the History of Logic, Vol. 8: The
Many-Valued and Non-Monotonic Turn in Logic, North Holland (Elsevier), Amsterdam-Oxford
2007, 129–204 (in particular pp. 132–137). Suggestions and historical recurrences aside, I think
that John is correct in recalling here the Aristotelian position against ECQ: in the Analytics,
several restrictions are in fact made on the acceptable form consequences must have. For
example, that the conclusion must be different from the premises, and dependent on them.
274 Chapter 10
must also recognize that several of these attempts are carefully organized and
well reasoned.
Finally, a few words on the use of Aristotle. John seems to assume that
Aristotle speaks a clear and entirely unambiguous language, and argues that
commentators have misrepresented Aristotle’s teaching because he considers
second substances as concepts. Apart from the correctness and consistency of
the interpretation, I would say based mainly on a passge of De Interpretatione
1, the curious fact is that all participants in the debate on universals ultimately
endorse John’s observation about the abundance of misrepresentations; of
course, they are always quick to point out others as those who pervert the text.
This charade is somewhat constrained by the literary genre of the authors,
which consists of commentary on authoritative texts, with all the rigidity that
such a format entails.
Nominal philosophers are wrong in not taking the Aristotelian realism into
due account:
7 Metal. 81, 20–26: ‘Alius sermones intuetur et ad illos detorquet quidquid alicubi de univer-
salibus meminit scriptum. In hac autem opinione deprehensus est Peripateticus Palatinus
Abaelardus noster, qui multos reliquit, et adhuc quidem aliquos habet professionis huius
sectatores et testes. Amici mei sunt, licet ita plerumque captivatam detorqueant litteram, ut
vel durior animus, miseratione illius moveatur.’
John of Salisbury on Universals 275
Another master deals with concepts, and says that only they can be con-
sidered as genera and species. They start from Cicero and Boethius, who
extol Aristotle as the originator of this opinion, since those things are
believed and said [by him] to be concepts. On the other hand a notion, as
they say, is the conceptual understanding of a certain thing which comes
from the perceived form, without interpretation. Elsewhere, they say that
a notion is a concept and a simple conception of the soul.9
John remarks that ‘what has been written [by Aristotle] deviates from the posi-
tion according to which the concept or the notion of universal expressions em-
bodies the universality.’10 The criticism of conceptualism or ‘notionalism’ ends
8 Metal. 81, 26–29: ‘Rem de re praedicari monstrum ducunt, licet Aristoteles monstruosi-
tatis huius auctor sit, et rem de re saepissime asserat praedicari, quod palam est, nisi
dissimulent, familiaribus eius.’
9 Metal. 81, 29–35: ‘Alius versatur in intellectibus, et eos duntaxat genera dicit esse, et spe-
cies. Sumunt enim occasionem a Cicerone et Boetio, qui Aristotelem laudant auctorem,
quod hae credi et dici debeant notiones. Est autem, ut aiunt, notio ex ante percepta forma
cuiusque rei cognitio, enodatione indigens. Et alibi: notio est quidam intellectus, et sim-
plex animi conceptio.’
10 Metal. 81, 35–36: ‘Eo ergo deflectitur quicquid scriptum est, ut intellectus aut notio univer-
salitatem universalium claudat’
276 Chapter 10
here, apparently with the usual summary judgment. The questions one could
ask about the scope of such a hasty judgment remain unanswered. Perhaps
the ‘deviation’ reported is a deviation from what Boethius actually says about
concepts.
The subsequent positions discussed are labelled as ‘realist’:
Those who believe that things inhere in other things have many different
opinions. Since everything that is is numerically unitary, [Boethius] con-
cludes that a universal thing either is numerically one or does not exist.
However, since it is impossible for universals not to be substantial things,
as the things of which they are substantial constituent do exist, they once
again relate universals to particulars as what must be tied to the essence.11
It is not clear (to me) how the argument’s antecedent ‘since it is impossible …’
connects to the rest of the reasoning attributed to realists. Namely, Boethius
states that everything that exists is numerically one and concludes that in
order to exist universals must be numerically one. At this point, a realist would
provide reasons to consider them unitary things. Instead, the realist invoked
by John contemplates the substantiality of universals. One can certainly take
this step, but it has little to do with the question at hand, unless additional as-
sumptions are made.
The first realist position examined by John is that of Walter of Mortagne’s
followers:
11 Metal. 81, 37–42: ‘Eorum vero, qui rebus inhaerent, multae sunt et diversae opiniones.
Siquidem hic, ideo quod omne quod <est> unum [est] numero est, aut rem universalem,
aut unam numero esse, aut omnino non esse concludit. Sed quia impossibile est substan-
tialia non esse, existentibus his quorum sunt substantialia, denuo colligunt universalia
singularibus, quod ad essentiam unienda.’
12 Metal. 81, 42–82, 47: ‘Partiuntur itaque status, duce Gautero de Mauritania, et Platonem,
in eo quod Plato est, dicunt individuum; in eo quod homo, speciem; in eo quod animal,
genus, sed subalternum; in eo quod substantia, generalissimum. Habuit haec opinio ali-
quos assertores, sed pridem nullus hanc profitetur.’
John of Salisbury on Universals 277
The style of the exposition is the same as the previous one; namely, a sketch of
the character and a rather sarcastic final judgment. Walter’s thesis is actually
rather elaborate, however, let me add some detail to complement John’s brief
account. Walter introduces and criticizes some claims that he considers realist.
In introducing his position, which can be labelled a ‘theory of (partial) iden-
tity’, he quotes Boethius’ maxims recalled by John: ‘omne quod est idcirco est
quia unum est’, ‘quicquid est, est individuum’; from these – with some implicit
premises – one can obtain the thesis that the same essence is individual and
universal. As we have seen, to say that Socrates is a species means that Socrates
is ‘equivalent’ to any other individual of that species, i.e. he is not different in
the specific status (for instance in being a man).
John’s distinction between nominal and realist philosophers is too sharp,
and does not allow for differentiation between realists who endorse strong po-
sitions in the style of William of Champeaux, and ‘weak’ realists such as those
who profess some form of (quasi-)nominalism, as Walter does. As for Abelard,
John takes into account only one aspect, which is perhaps the best known of
the doctrines that he presents. Walter would have subscribed to the thesis that
universals equal individuals, but he also would have noted that his position
is the exact opposite of realism. The title of the chapter mentioned above re-
fers to the different positions on species and genera. In his criticism of realists,
John certainly groups everyone together, but not without reason, because all
the positions we have examined in the course of this work are ‘compromised’
by some form of realism.
After Walter comes Bernard of Chartres, who certainly occupies more of the
text than any others (he receives almost a column of Patrology, while others
are discussed in a few lines at most). John begins by recalling the position of
Bernard and his followers on ideas:
He13 pays attention to the ideas as if to emulate Plato and imitate Bernard
of Chartres, and says that species and genera are no more than those. An
idea, as Seneca says, is the eternal exemplar of things generated by na-
ture. Since universal things are not subject to the processes of corruption,
neither are they altered by the changes that affect individual things […].
13 ‘Ille …’. Some interpreters have wanted to connect the pronoun to the character that pre-
cedes it, i.e. Walter of Mortagne (cf. P.E. Dutton, cit., 213; D.D. McGarry, in The Metalogicon
of John of Salisbury, University of California Press, Berkeley – Los Angeles 1962, p. 113
n. 284). This reading seems natural, but has the disadvantage of attributing a form of ‘ide-
alism’ to Walter, as well as the position on the status (in contrast, I would say, to an idealist
doctrine in Platonic style).
278 Chapter 10
They are said to be universal in a strict sense, given that individuals are
considered unworthy of being denoted by the substantive verb [i.e. are
not among things properly denoted by ‘being’].14
The reference is to Seneca, Epistole 58.19, one of the Platonic streams that
flow into the (late) Middle ages. Boethius, quoted shortly afterwards, speaks
of ‘beings’ in the specific sense of De Inst. Arithmetica, De Trinitate and Contra
Eutychen. The De Divisione Naturae represents another alleged source. Scotus
Eriugena shares the division between things that are (in a positive sense) and
things that are not (in a negative sense), or vice versa.15
Among beings (esse) ‘which do not allow increase or decrease’, John-
Bernard lists quantities, qualities, relations, places, times, shapes, and ‘every-
thing which is connected to bodies’. Apart from forms, perhaps placed between
things ‘related to bodies’, the partially non-Aristotelian classification of things
in the sense proposed by Boethius and Nicomachus is put forward again.
According to Bernard, these are immutable, whereas individuals are subject
to the processes of generation and corruption. Here John quotes Heraclitus’
famous assertion that ‘we descend and do not descend into the same river’,
explaining that waves constantly change, while ‘the movement in the river re-
mains unchanged’.16
This exposition of Bernard’s doctrine continues with some further consid-
erations on ideas:
14 Metal. 82, 47–55: ‘Ille ideas ponit, Platonem aemulatus, et imitans Bernardum
Carnotensem, et nihil praeter eas, genus dicit esse vel speciem. Est autem idea, sicut
Seneca definit, eorum quae natura fiunt exemplar aeternum. Et quoniam universalia cor-
ruptioni non subiacent, nec motibus alterantur, quibus moventur singularia […] proprie
et vere dicuntur esse universalia, siquidem res singulae verbi substantivi nuncupatione
creduntur indignae.’
15 The same quote from De Inst. Arithmetica is present in Metalogicon and De Divisione.
Some twelfth century manuscripts contain fragments of De divisione naturae; one that
is particularly interesting to me (without exaggerating its importance) is the Oxford
Bodleian Library Auct . F.3.15, where a series of extracts from Scotus’ work are connected
to the Timaeus and Bernard’s Glosses. The text was studied by I.P. Sheldon-Williams (‘An
Epitome of Irish Provenance of Eriugena’s De divisione naturae (MS. Bodl. Auct. F.3.15)’,
Proceedings of the Royal Irish Academy 58, Section C (1956), 1–16) and referenced by
P.E. Dutton (op. cit., 207).
16 Seneca is quoted in the critical apparatus of the De Divisione Naturae but not as a source
for the quotation from Heraclitus, which is referenced in the sixth letter to Lucilius
(Ep. 58.23). In any case, the texts share the use of evocative metaphors as a supplement to
the dogmatic exposition.
John of Salisbury on Universals 279
These ideas, i.e. exemplary forms, are the primitive reasons of all things;
they are not subject to decrease or increase, are stable and perpetual, so
that even if the material world should perish, they would not. The series
of all corporeal things is preserved by means of these, as St. Augustine
seems to maintain in De libero arbitrio, since these always are; while tem-
poral things may cease to exist, the number of things does not decrease
nor increase.17
This passage is interesting for at least two reasons. One – a logical reason – has
to do with the theory of meaning. I do not know if John, or the thinkers he re-
fers to, considered it, but some authors (including Walter and Abelard) speak
of the difficulty of seeing denoted things as the principal meaning of the terms
due to their instability. The other one – a metaphysical reason – can be linked
to Boethius, who distinguishes between immutable essences and mutable in-
dividual bodies.
John’s conclusion, in his typical fashion, is simultaneously admiring and
ironic.18 The doctrine of ideas, he states, is a great thing for refined intellects
but has nothing to do with Aristotle; Bernard’s followers have worked hard at
putting together Platonic and Aristotelian doctrines, but have arrived too late
to the debate and work in vain in an attempt to reconcile dead thinkers who
disagreed while they were alive.
The last representative of the bad interpreters of Aristotle that John dis-
cusses is an unnamed follower of Gilbert of Poitiers:
17 Metal. 82, 69–76: ‘Haec autem ideae, id est exemplares formae, rerum primaevae omnium
rationes sunt, quae nec diminutionem suscipiunt nec augmentum; stabiles et perpetuae,
ut etsi mundus totus corporalis pereat, nequeant interire. Rerum omnium corporalium
numerus consistit in his; et sicut in libro De libero arbitrio videtur astruere Augustinus,
quia hae semper sunt, etiamsi temporalia perire contingat, rerum numerus nec minuitur,
nec augetur.’
18 John’s style has been studied in several works, including C.S. Jaeger, ‘Irony and Role-
Playing in John of Salisbury and the Becket Circle’, in M. Aurell (ed.), Culture politique
des Plantagenêt, 1154–1224 (Actes du colloque tenu à Poitiers du 2 au 5 mai 2002), CESCM,
Poitiers, 2003, 319–331.
280 Chapter 10
The passage quoted refers to the important issue of the relationship between
samples and exemplars. John speaks of particular native forms, and some in-
terpreters have used this text to support the thesis of a kind of ontological
particularism, so that only particular forms would exist for Gilbert. However,
the thesis – which is not directly attributed to Gilbert – contemplates samples
and exemplars. John says that Gilbert’s follower asserts the conformity of these
forms, which seems to indicate that he attempts to join universality and par-
ticularity, and/or define the network of relations connecting universal and par-
ticular things.20
In conclusion, John recalls the Boethian position on universals: ‘perceivable
in what is perceived by senses, but conceived by the mind as non-perceivable,
particulars in particular things, but universals in everything’, a position in
the same vein as the neoplatonic doctrine of the universal’s threefold way of
being. In his commentary on De Trinitate, Gilbert speaks of the different types
of intellectual abstraction and their proper objects. Somehow this epistemic
perspective is complementary to the ontological one of the second Boethian
commentary: in the latter the universals’ mode of being is considered, and in
the De Trinitate the way of thinking about universals and other abstract enti-
ties is object of theoretical investigation.
‘Est et alius’ which, following the teaching of Joscelin of Soissons, attributes
universality to things ‘in unum collectis’. In Chapter 7 we examined the De
Generibus et Speciebus (attributable to Joscelin). This treatise contains a section
in which the author deals with the notion of whole, another where certain the-
ories on universals are criticized, and a final section where the solution to the
problem of genera and species is introduced. Joscelin’ position is ambivalent:
he considers on the one hand collections of individual essences as meanings
of universal expressions, on the other he considers totalities formed by atomic
material parts. In both cases – though not without difficulty – the relation of
19 Metal. 83, 84–91: ‘Porro alius, ut Aristotelem exprimat, cum Gilberto episcopo Pictavensi,
universalitatem formis nativis attribuit, et in earum conformitate laborat. Est autem
forma nativa originalis exemplum, et quae non in mente Dei consistit, sed rebus creatis
inhaeret. Haec Graeco eloquio dicitur idos, habens se ad ideam, ut exemplum ad exem-
plar; sensibilis quidem in re sensibili, sed mente concipitur insensibilis, singularis quoque
in singulis, sed in omnibus universalis.’
20 We have mentioned the difficulty of this task in the chapter devoted to Gilbert.
John of Salisbury on Universals 281
The same irony is addressed to those who use a ‘new language due to having
no familiarity with Latin.’ These authors claim that universals are maneries of
things. John says he does not know the origin of this expression, but suggests
that it is used by the author of Glossemata or a contemporary. According to
John, the term may refer to a collection of things or to the status for which
those things subsist (Abelard in fact employs this expression in the Logica
Ingredientibus in the latter sense).
The exposition of the realist theses on universals concludes with a brief ref-
erence to an independent position on the status: ‘there are even those who, in
considering the status, say that genera and species are statuses.’22 It is unclear
if John thinks that there is any connection between the three positions quoted
on the status rerum. Abelard and Walter have their own opinion on the subject,
but after having said that Walter is a supporter of the doctrine of status, John
subsequently makes no mention of him.
21 Metal. 83, 94–96: ‘Cum ad interpretandas auctoritates ventum est, laborat prae dolore,
quia in locis pluribus rictum litterae indignantis ferre non sustinet.’
22 Metal. 83, 106–107: ‘Nec deest qui rerum status attendat, et eos genera dicat esse et species’.
282 Chapter 10
What would it mean to ask what the genus is if one thought it were abso-
lutely nothing? Of what is not, it makes no sense to ask what, how much
or how it is.23
John had just claimed that he wanted to provide a faithful reading of Aristotle;
here he uses Boethius, who is commenting on Porphyry, to introduce the con-
ceptualist position, in a somewhat convoluted way. The best thing he could do,
at least initially, is to provide textual evidence, which proves, however, to be
rather difficult. Instead, he takes his cue from a text by Boethius, and argues
that the categories are useless if Aristotle is right in considering universals to
be non-existing things.24 Thus, John wants to formulate the following thesis:
by saying that genera and species do not exist, Aristotle denies their existence,
but not their mental being.
Authors prior to John do not take seriously the conceptualist option.
Boethius introduces it, but is more interested in that to which the concepts
refer than the concept as such; it is not the objects in the mind that must be
considered, but the objects of, or for, the mind. He then goes on to make sev-
eral interesting assertions about the process of abstraction. John utilizes this
psycho-epistemic aspect of the Boethian commentary, but fails to offer an ex-
planation. In fact, Boethius places the thesis of non-existence of genera and
species in the introduction to his remarks on Porphyry, without attributing it
to Aristotle. In any case, John is convinced that the thesis is Aristotelian, and
asserts that an interpreter of the Categories has the choice between either dis-
agreeing Aristotle or discarding all that has been said (between the eleventh
and twelfth century) on the issue of universals.
John paraphrases the Boethian text on the process of abstraction thus:
We should not fear that the intellect is ineffective, the intellect which
perceives that which cannot exist separately, deriving it by abstraction
from particular things. […] Afterwards, the intellect sometimes considers
a simple thing as it is; for example Plato, sometimes in another mode, by
composing things that are not composed, sometimes abstracting those
that cannot be separated from their objects.25
23 Metal. 85, 4–6: ‘Quid ergo ad rem attinet quaerere quid genus sit, si illud omnino non esse
constiterit? In eo enim quod non est inepte quaeritur quid, quantum aut quale est.’
24 Metal. 85, 79: ‘Ergo si Aristoteles verus est, qui eis esse tollit, inanis est opera praecedentis
investigationis’
25 Metal. 85, 17–86, 27: ‘Nec verendum ut cassus sit intellectus, qui ea percipit seorsum a
singularibus, cum tamen a singularibus seorsum esse non possint. […] Porro simplex rem
John of Salisbury on Universals 283
So far, John has followed Boethius quite closely: the description of the abstrac-
tive process is intended to prove that the abstracting intellect separates form
from matter by considering it something that does not exist separately from
the reality in which it is located. At this point, John begins to declare his onto-
logical preferences:
There is only one way in which things are, the one allowed by nature, but
it is not the only way to understand or mean: though there is no man who
is neither this or that, and furthermore, something can be thought of or
meant in the way in which this or that cannot be thought of or meant.
So, genera and species are conceived by the abstracting intellect in order
to signify simple entities; those who seek them in nature as objects of
perception, labor in vain, since in fact nature has not generated any such
thing.26
There are no particularly innovative theses in this passage: many of the pro-
tagonists of the debate on universals would agree in claiming that only physi-
cal objects exist in nature. The allusion to what can be thought of/meant still
awaits clarification: the objects of thought or meaning can be considered ei-
ther internal to the mind or independent of the mind, and perhaps belonging
to a domain of entities that ‘are’ in a broad sense of the term.
John recalls another aspect of Boethius’ doctrine; namely, that reason grasps
the similarities of things ‘by placing them before itself’. The problem here is
to understand the relationship between similarities as objects of thought and
things (remember that according to Boethius, similarities subsist in a certain
way and are thought of in another way). The following text deals with this
point:
Thus, genera and species are not entities separated from singular things
in act or by nature, but certain mental images of natural and actual
interdum inspicit ut est, ut si Platonem attendat, interdum alio modo, nunc enim com-
ponendo quae non sunt composita, nunc abstrahendo quae non possunt esse disiuncta.’
26 Metal. 86, 41–49: ‘Et quidem rebus existendi unus est modus, quem scilicet natura contu-
lit, sed easdem intelligendi aut significandi non unus est modus. Licet enim esse nequeat
homo, qui non sit iste, vel alius homo, intelligi tamen potest et significari, ita quod nec
intelligatur nec significetur iste vel alius. Ergo ad significationem incomplexorum, per
abstrahentem intellectum genera concipiantur, et species, quae tamen, siquis in rerum
natura diligentius a sensibilibus remota quaerat, nihil aget et frustra laborabit, nihil enim
tale natura peperit.’
284 Chapter 10
The soul indeed, which has become almost a mirror for its speculative
power, finds what it defines in itself. In fact, the exemplar is in it, and the
example in the actual things. Like in grammar, when it is said: ‘nouns that
have this ending are feminine or neuter,’ a paradigm is given, a model
for many declinable nouns […] thus, some exemplars are thought of by
the mind, while examples are molded and supplied to the senses by na-
ture. These specimens are objects of thought (cogitabilia) and are almost
fantasies or shadows of existing things; according to Aristotle […] they
are things that reveal themselves to (monstra) and concern the intellect
alone.28
27 Metal. 86, 54–59: ‘Sunt itaque genera et species non quidem res a singularibus actu et
naturaliter alienae, sed quaedam naturalium et actualium phantasiae renidentes in intel-
lectu, de similitudine actualium, tamquam in speculo nativae puritatis ipsius animae […]
hoc est rerum imagines in mente apparentes.’
28 Metal. 86, 59–87, 72: ‘Anima enim quasi reverberata acie contemplationis suae, in se ipsa
reperit quod definit; nam et eius exemplar in ipsa est, exemplum vero in actualibus. Sicut
cum in grammatica dicitur: ‘nomina quae sic desinunt feminina vel neutra sunt’, gene-
ralis quaedam praescribitur ratio quae quasi multorum declinabilium exemplar est […]
sic quaedam exemplaria concipiuntur in mente, quorum exempla natura formavit, et
sensibus obicit. Illa itaque exemplaria cogitabilia quidem sunt et sunt quasi phantasiae
et umbrae existentium, secundum Aristotelem […] monstra enim sunt et soli intellectui
patent.’
John of Salisbury on Universals 285
This text contains many implications as well as several problems. Let us start
from the example/exemplar distinction. One source may be the medieval neo-
platonic tradition,29 even if some anomalies are present. An element worth
noting is the type of example that John chooses to illustrate his position.
Grammarians do not speak of specific grammatical expressions, but of type-
expressions and paradigmatic cases. This way of putting things moves the dis-
cussion away from the neoplatonic context, and indicates similarities to the
type-token distinction familiar to us. John’s analysis, although not developed,
proposes an interesting point of view not shared, to my knowledge, by other
participants in the contemporary debate on universals.
One may wonder if the indecision about the meaning of Boethius’ simi-
larities is reproduced on another level. It is not easy to answer: the term ‘cogi-
tabilia’ could refer to objects of thought, possibly independent of it, while
the next elucidation ‘almost fantasies and shadows … related to the intellect
alone’ places such items back into the mind which thinks of them. The fact
that John talks of ‘shadows of existing things’ is rather odd: in the neoplatonic
tradition we have mentioned, physical objects, and possibly concepts, are the
shadows of ideas; here John, at least partially, overturns the traditional picture
of exemplarism.
Following the well-established habit of his teachers, our author offers a re-
reading of the authoritative texts from his own point of view. The following
passage is particularly significant, also because of its resemblance to passages
of the Logica Ingredientibus:
29 Among other sources, Seneca (whose important sixth letter to Lucilius is mentioned
by John), St. Augustine, Chalcidius (as translator and commentator of the Timaeus),
Boethius, Scotus Eriugena, William of Conches, Bernard, and Theodoric of Chartres. The
following text by Bernard (most likely known to John) is one of the many examples: ‘mun-
dus factus est ad exemplum immutabile. Ergo est imago alterius, id est imago archetipi.
Dicunt quidam exemplum esse ad cuius similitudinem aliquid fit, exemplar quod ad alte-
rius fit […] alii e converso.’ (Glosae super Platonem, ed. cit., 161, 109–112).
286 Chapter 10
granted that universals are or are things, if that is how obstinate people
like it; the number of things does not increase or decrease because of this,
as these entities are not to be enumerated among existing things.30
Abelard says something quite similar regarding the main meaning of common
nouns.31 The shifting from signification to conception does not definitively
mean that John holds a conceptualist position; as usual, the alternative is to
give importance to the mental sphere or the ‘things’ to which concepts refer, in
this case the causes of conception and nature. The final reference to ontologi-
cal parsimony (and/or ontological tolerance) does not help to clarify in what
direction we should move. The swing between psychological reduction and
ontological foundation continues in the explanations that follow the quoted
text. The best I can do is to present these two movements, and allow the reader
to draw his or her own conclusion.
The need for an ontological foundation arises in statements such as:
‘apart from singular things there is nothing; perhaps in addition to individual
things, true qualitative objects (qualia) exist, or similar meanings of complex
expressions.’32 It is unclear what John means by qualia, but everything leads
me to overlap this quasi-concession with Abelard’s doctrine on non-things
30 Metal. 87, 72–88, 101: ‘Quod autem universalia dicuntur esse substantialia singularibus,
ad causam cognitionis referendum est, singulariumque naturam; hoc enim in singularis
patet, siquidem inferiora sine superioribus nec esse nec intelligi possunt. […] Universalia
tamen et res dicuntur esse […] Nam et ea quae iacent sub affirmatione et negatione res di-
cuntur, et vera sepissime dicuntur esse, nec tamen substantiis vel accidentibus aggregan-
tur, aut nomen suscipiunt Creatoris aut creaturae […] in philosophantium foro exuberat
bonitas, et verba distrahuntur ad gratiam. Itaque detur ut sint universalia, aut etiam ut
res sint, si hoc pertinacibus placet; non tamen ob hoc verum erit rerum numerum augeri
vel minui pro eo quod ista non sunt in numero rerum.’ I have not accept the integration
introduced in the critical edition in the final sentence: ‘augeri vel minui pro eo quod ista
non sunt <vel sunt> in numero rerum.
31 I shall mention just one among many textual evidences: Log. Ing., 29, 11–15: ‘et dicun-
tur universalia subsistere in sensibilibus, id est significare intrinsecam substantiam
existentem in re sensibili […] eandem tamen naturaliter separatam a re sensibili demon-
strant’; Log. Ing. 84, 23–25: ‘et quod (dicitur) res speciei non posset esse sine differentia,
hoc secundum communem causam impositionis nominis universalis accipiendum est’;
Log. Ing., 45, 28–30: ‘Boethius tamen praedicari de pluribus largius accipiens haec quoque
universalia dicit secundum communem impositionis causam.’ We must remember that the
common cause of imposition is the status, the primary meaning of a common noun.
32 Metal. 88, 110–112: ‘… seorsum a singularibus nihil est, nisi forte qualia sunt vera, aut si-
milia complexorum significata sermonum.’
John of Salisbury on Universals 287
(1) Anyone who knows good things and practices them is wise and happy
(2) I promise a horse. Is the horse I promise healthy or sick?
(3) The woman who condemned, saved
(4) The wood that gave death gives life
288 Chapter 10
At least two of these statements are ambiguous, but this is not the central
point for John; the discussion focuses on the generic reference of nominal
expressions and the use of demonstrative pronouns.33 This theme is inserted
immediately after the passage quoted above, in which John proposes a broad
definition of a universal meaning (‘what an expression true of several things
commonly means’). We must recall that John’s point is a) there are only in-
dividual existing things, b) it is acceptable to talk in a broad and perhaps
metaphorical sense of universal things, c), items such as qualia or complex
meanings should perhaps be included in the ontological inventory.
John observes that in the sentence ‘wise, happy man is he who knows good
things and practices them’, despite not having a defined reference, the pro-
nouns ‘who’ and ‘them’ give rise to some restriction of ‘universality’ through
a determination of the act of knowing. What John seems to be saying is that
there is no particular object to which the pronouns refer, since the sentence
speaks of a wise man and generic good things. The emphasis is not placed on
the reference of the phrase ‘wise man’ or ‘good things’, but on the role of the
pronouns. Predicate logic can be helpful in interpreting the text: In the transla-
tion of natural language into that of first order logic, variables typically take
the place of pronouns and their intended meaning is represented by elements
of the domain of interpretation. In the cases considered, a translation of sen-
tence (1) might be:
∀x∀y. ((y) is a good thing ∧ (x, y) knows & practices) → (x) is a wise happy
man
33 A vast amount of literature is available on the subject, from medieval texts before and
after the theory of suppositio, to contemporary discussions on anaphora from both a
linguistic and a logical-philosophical point of view. Among the numerous studies, I just
mention the work by C.H. Kneepkens, ‘Mulier Quae Damnavit, Salvavit. A Note on the
Early Development of the Relatio Simplex’, Vivarium 1976 (XIV, 1), 1–25, and the fine article
by R. Hülsen, ‘Understanding the Semantics of ‘Relativa Grammaticalia’. Some Medieval
Logicians on Anaphoric Pronouns,’ in K. Von Heusinger – U. Egli, Reference and Anaphoric
Relations, Kluwer, Dordrecht 2000, 31–46.
John of Salisbury on Universals 289
where ‘promise’ is a ternary relation, whose arguments are who promises (a),
the thing promised (x) and the person x is promised to (b). The reading returns
the same kind of problem: there is nothing in the domain of interpretation
which verifies the relation of promising, since he who promises does not com-
mit himself to any particular horse. There are at least two possible attitudes in
such cases: the first is to consider verbs such as ‘promise’ different from other
transitive verbs, in requiring objects of a special sort as arguments; the second
is to make the translation more complicated in such a way as to allow the or-
dinary assignment of truth-values. Both perspectives have supporters in late-
termism and among philosophers closer to the present day.
Abelard, who discusses similar cases, slightly wavers in his position: on the
one hand he seems to accept the idea that non-discrete things may appear as
arguments of ‘intensional’ verbs; on the other he believes that the sentence can
be reformulated in such a way as to be ontologically safe, in some nominalist
sense. John recognizes that verbs such as ‘promise’ are of a particular type, but
does not seem to accept the idea that their argument is equally special. The
text is extremely brief and difficult to interpret:
The species, that is the discrete thing, does not derive its obligation from
the genus. When I say ‘it is promised’, ‘meant’, ‘understood’, and the like,
something discreet is the subject of the promise or signification, but only
provided that the relation is regular.34
34 Metal. 89, 139–143: ‘Species, id est res discreta, obligationem generis non attingit. Cum
enim dico quod promittitur, quod significatur, quod intelligitur et similia, discretum ali
quid promissioni, aut significationi subjectum est et si tamen propria est ipsa relatio.’
290 Chapter 10
35 Or ‘non-specific reference’. My belief stems from the fact that I cannot make sense of
an interpretation in which the promise relates to the species ‘horse’ or genus ‘animal’.
In addition, John says, ‘the species, that is the discrete thing.’ The English translation by
D. McGarry is along the same line (The Metalogicon of John of Salisbury: A Twelfth Century
Defense of the Verbal and Logical Arts of the Trivium, University of California Press,
Berkeley, 1955; rep. pap. Paul Dry Books, Philadelphia, 2009). Contrary to my reading,
however, one could cite the note introduced by a ‘puta’ in which John explains that the
species is something like /man/ and genus something like /animal/, and these ‘things’ are
properly objects of signification or conception. This observation is linked to a theory on
meaning which we will consider in the next paragraph.
36 The term ‘improper’ does not appear in the text; I have used it as opposed to relatio pro-
pria, which is used for intensional verbs. Relations like the ones signified by verbs in 3
and 4 are somewhat enigmatically called ‘relationes in genere quae, salvo intellectu veri,
nequeant revocari ad speciem.’
37 Indeed, there are several ways forward as well as several problems; I cannot here deal
even superficially with issues related to intensional verbs and generic reference, so I
shall limit myself to a marginal consideration on a passage of the Logica Ingredientibus.
In commenting on the example ‘mulier quae damnavit, salvavit’ Kneepkens considers
whether ‘eadem secundum indifferentiam sexus’ means something like ‘the same secun-
dum indifferentiam sexus.’ The problem with this reading is that it brings Abelard too close
to the theory of non-difference, which he harshly criticizes. There are two possibilities: 1)
Abelard accepts some postulates of the theory of non-difference, without accepting that
the same theory could be useful in solving the Porphyrian problem; the reading would
be: ‘one woman saved and another one (different but linked by a relationship of non-
difference) damned’. 2) ‘eadem secundum indifferentiam sexus’ should read ‘‘the same’
(is said) about the non-difference of sex’. The example above would equate to ‘a woman
damned and one of the same sex saved’, where the burden of reflexivity is moved onto sex.
What we read in Glose Admirantes (mid- thirteenth century) is interesting in this respect:
the author – quoted by Kneepkens (ibid., 3–4) – argues that in the sentence ‘mulier quae
damnavit salvavit’ the relative pronoun ‘quae’ cannot refer to the same person, nor to the
species, since in both cases the statement would be false; rather it must be understood in
John of Salisbury on Universals 291
that, if the truth and univocity of the sentences are maintained, the objects of
discourse are to be considered generic; obviously, choosing this path one has
some explanation to give about the fact that such entities admit contrasting
properties. John apparently feels no obligation to do so, nor has he anything
to say about the logic of the generic entities; he merely observes that in these
cases the expressions ‘do not refer to the species but remain with the genus.’
Metalogicon Dialectica
a mixed way, ‘as argued by ancients’: a woman damned and one of the same species saved.
This interpretation coincides perfectly with alternative (2).
292 Chapter 10
(cont.)
De Intellectibus
It is usual to ask about the signification and the conceptual meaning of universal
expressions, what they mean, or what is included in their concepts […] They say that
just as when one has a sensory experience of a man, he necessarily perceives this
or that […]. One who has a man in mind has this or that in mind; that appears to
be completely false. […] When you say, then, that you think of a man, the sense is
that by means of the intellect someone has the human nature in mind, that is, the
differentiated animal ( =the rational mortal animal)’.c
a Metal. 89, 151–161: ‘Homo animalium species est […] animal genus rerum est. Est enim signifi-
catum vocis, in quo descendit, aut quod animus rationabiliter concipit ex voce audita. Itaque
qui audit hanc vocem homo, nec omnes percurrit homines (quoniam hoc infinitum et vires
excedit) nec tenetur in uno […] Sic et qui definit animal esse substantiam animatam atque
sensibilem non unum aliquid ne imperfectum, nec omne definit, ne laboret in infinitum.
Singulum enim horum non simpliciter quid significat, aut definit, sed potius quale quid.’
b Dial., 112–13: ‘Alii enim omnia quibus vox imposita est, ab ipsa voce significari volunt, alii vero
ea sola quae in voce denotantur atque in sententia ipsius tenentur […] Manifestum est eos
velle vocabula non omnia significare quae nominant, sed ea tantum quae definite designant,
ut ‘animal’ substantiam animatam sensibilem […] (‘non-homo’) ea quae significare potest
infinita sunt […] si tamen significare proprie ac secundum rectam et propriam eius definitio-
nem sumamus, non alias res significare dicemus nisi que per vocem concipiuntur’.
c De Intellectibus (attributed to Abelard), edited in L.U. Ulivi, La psicologia di Abelardo e il
‘Tractatus de intellectibus’, Edizioni di storia e letteratura, Roma 1976, 123 22–124, 2.: ‘Solet
frequenter quaeri de significatione atque intellectu universalium vocum, quas res videlicet
significare habeant, aut quae res in eis intelligatur […] Sicut enim inquiunt, cum homo sen-
titur necesse vel hunc vel illum […] qui hominem intelligit, profecto quemdam hominem
intelligit, et ita hunc vel alium, quod omnino falsum apparet. […] Cum itaque dicitur ‘homo
intelligitur’, hic est sensus, quod aliquis per intellectum naturam concipit humanam, hoc est
animal tale attendit.’
These texts seem to say the same things, although in a slightly different way.
For Abelard, what is mainly signified by generic-specific terms are not individ-
ual things, but status, esse or common causes. For John, similarly, what is signi-
fied are not individual things, the individual quids, but qualified quids named
by defining nominal clauses.
The quale quid may act as a substitute for the quid(s) in sentences like (3)
and (4), in which the subject or pronouns refer to a generic object. This is con-
firmed by the text immediately following the clarifications on the meaning of
universal terms:
John of Salisbury on Universals 293
In a similar way, in the Tekne Galen says that medicine is the science of
things that are healthy, sick or neither healthy nor sick. He does not speak
of all things, as this would require infinite enumeration, nor of some of
them, because it would not be theoretically correct, but rather of quali-
fied quids. Similarly, Aristotle says that genera and species determine a
quality about substance, not just a quid, but somehow they determine a
quale quid.38
The qualified quids are evidently the meanings of the expressions ‘healthy
thing’, ‘sick thing’, and ‘thing neither healthy nor sick’. The reference to the
Aristotelian text apparently serves the purpose of providing the reader with an
independent reason to accept the terminological innovation. Aristotle’s text is
often quoted, but interpreted differently from the beginning of the twelfth cen-
tury onwards. It is usually compared to the passage by Priscian in which nouns
are defined as expressions signifying substance and quality. Abelard deals with
these texts at length, and interprets them according to his semantic theory:
a universal term signifies substance in the sense that it denotes one or more
individual substances, and signifies a quality because it has a semantic relation
of determination (or meaning in the strict sense) with a certain property.
Abelard himself seems to distinguish between the determinatio and the
main impositive meaning. In other words (if I understand correctly), a term
such as ‘homo’ determines the different properties of rationality, mortality etc.,
but has as its main impositive meaning /mortal rational animal/, that is, the
status of man. Somehow the notion of quale quid introduced by John does not
imply a single quid object of naming, nor a quality, but a qualified quid, in
correspondence with Abelard’s main impositive meaning. John also seems to
share my hesitation to match determinative and main impositive meanings
(due perhaps to the difficulty in obtaining a definite answer from Abelard’s
teaching). Hesitation aside, John’s determinative meaning seems to coincide
perfectly with Abelard’s main impositive.
The text above constitutes a kind of boundary between Abelard’s doctrine
and John’s. The line is not crossed, however; on the contrary, from this point
of almost coincidence, a long retreat begins which spans four columns of the
Patrology. In distancing himself from Abelard’s position, John notes that he
does not really understand what a non-determined object is, nor what the
38 Metal. 89, 162–90, 2: ‘Simile est quod Galienus in Tegni (= art. med. 2, ed. Kuhn, 1.30) me-
dicinam dicit scientiam sanorum, aegrorum, et neutrorum. Non inquit omnium, quo-
niam hoc infinitum, non quorundam, quoniam hoc ad artem imperfectum, sed potius
qualium quorum. Item Aristoteles: genera, inquit, et species circa substantiam qualita-
tem determinant, non enim simpliciter quid, sed quodam modo quale quid determinant’
294 Chapter 10
status is. The domain of things is fixed by nature once and for all, with only
existing discrete things belonging to it. According to John, this assertion does
not go against what is usually said about the difference between meaning and
denotation, provided that one intends ‘the simple relation’ for meaning. No
other explanation is provided, and the text returns to the theme of the demon-
strative and relative pronouns.39
John tries to show that no spurious entity is present in the catalog of created
things. Special attention is given to the subject of matter and form, and the
exegesis of biblical texts. I will consider just a few passages:
As Augustine says, God created matter molded in such a way that it does
not exist in any way without form, despite the fact that in some places it
is said to be ‘formless’.40
Form exists for the matter, just like matter is shaped by the form. In
fact, form is not an existing thing in itself (res existens) nor is matter
something discrete without the contribution of form.41
Though science, which considers things in abstract, and by its preci-
sion separates what is united by nature, […] one thing cannot subsist
without the other (i.e. form without matter).42
John envisages the creation of natural things as an instantaneous act: God did
not create form and then shape the matter, rather, He created entities already
formed. There is no need of any principle of individuation, since there is no
matter that takes form nor a form that adheres to shapeless matter. However,
Genesis seems to make explicit reference to generic or formal entities:
Heaven and earth are mentioned as the first created things […]. The bibli-
cal formula refers to the species of things, not of those things that logi-
cians imagine as not subordinate to the Creator, but the forms in which
things in their essence were produced at the beginning […]. Indeed, what
39 John is apparently looking for a good theory of distribution that allows him to explain the
grammatical role of the reflexive in senteces like ‘omnis homo diligit se’.
40 Metal. 92, 271–273: ‘Ut enim ait Augustinus, formatam creavit Deus materiam, siquidem
etsi alicubi dicatur informis, nunquam omnino existit sine formis.’
41 Metal. 93, 287–289: ‘Per materiam existit forma, sicut discernitur materia ipsa per for-
mam. Nam nec forma res per se existens est, nec materia sine formae beneficio discreta
est.’
42 Metal. 93, 302–306: ‘Etsi mathesis, quae doctrinaliter in abstractionibus versatur, et sub-
tilitate sua, naturaliter coniuncta disiungit […] tamen alterum sine altero esse non potest.’
John of Salisbury on Universals 295
John believes that his exegesis serves to demonstrate that genera and species
were not created by God, and therefore do not exist. Whether this argument
is convincing or not, it is to be noted that the realism at stake has to do with
forms, i.e. the position according to which forms exist independently of mat-
ter. Among the various realist or quasi-realist theories, only some take on a for-
mal point of view on universals. It must also be noted that the retreat from the
theory of status is made not by contesting what Abelard says (in fact, John only
says he does not really understand the status theory), but by criticizing a form
of realism certainly not shared by the peripatetic Palatine. The conclusion of
the scriptural investigation is as follows:
In De sex dierum operibus [by Thierry of Chartres] all good things created
are remembered and listed under their genera, but there is no mention
of universals […]. Thus, universal things are merely objects of thought,
in agreement with Aristotle, but there is nothing universal in the act of
things. [Genera and species] can be understood as metaphorical repre-
sentations, and their names are introduced by convention into a theoreti-
cal context. In fact, every man is this or that, i.e. a singular thing, but as
one can think of man in such a way so as not to think of a this or that man
[…], the thing thereby understood, even if is not understood by anyone,
is said to be common.44
43 Metal. 93, 308 94, 318: ‘Unde ab initio creata memorantur coelum et terra […] et in ipsa
expressione rerum habita est mentio specierum, non illarum dico quas logici fingunt non
obnoxias Creatori, sed formarum in quibus res prodierunt primo in essentiam suam […]
nam hoc ipsum quod aliquid coelum aut terra dicitur, formae effectus est. Item produxit,
inquit, terra herbam virentem, et lignum tale et tale, ut materiae formas doceret esse
concretaa, et Deum, sicut herbae, ita et viroris auctorem.’
44 Metal. 95, 364–378: ‘In operibus sex dierum in genere suo bona singula creata memoran-
tur, nec tamen creationis universalium mentio aliqua facta est […] ergo dumtaxat intelli-
guntur, secundum Aristotelem universalia, sed in actu rerum nihil est quod sit universale.
A modo enim intelligendi figuralia haec, licenter quidem, et doctrinaliter, nomina indita
sunt. Nam omne quod homo est, iste vel ille est, id est res singularis est; sed quia intelligi
potest homo ita quod nec iste intelligatur, nec ille […] res quae sic intelligi potest, etsi a
nullo intelligatur, dicitur esse communis.’
296 Chapter 10
Back to the starting point John finds himself having to face Boethius’ warning
about the objective correlate of concepts. The ‘common thing understood’ in
the text is characterized as a conformitas deducta rerum cogitatione, a confor-
mity derived from reflection on things. This idea of conformity perhaps un-
derlies the concession made above about the meaning of universal terms, and
invokes Boethius’ notion of similitudo. The idea is developed in the following
propositions:
Propositions (1) and (2) are rather insidious for two reasons at least: a) a simi-
larity or conformity depends on the attribution of properties;45 b) John’s com-
ments about the lack of scriptural texts on universals may return to haunt the
author. How do similarities rank in the catalog of created things? In (3) John
mentions forms again. In this case it is probably a matter of individual forms,
as only an individual form can be similar to another form. Against this appar-
ent evidence is the fact that the form John speaks of is called ‘species’; does he
perhaps have some sympathy for the identity theory? This is unlikely, since it
is among the theses criticized in the pars destruens of Metalogicon. In (4) John
passes without any explanation from the form to the image common to several
forms. The term ‘imago’ is ambiguous: on the one hand it refers to a mental
dimension, and on the other it suggests what Boethius and others say about
forms in their state of secular degradation.
45 Let us consider for example the symbolic writing: aShb ↔ h(a)∧h(b). The relation of simi-
larity depends on the parameter h. In other words, we do not save properties by introduc-
ing relations. Some contemporary theories consider the similarity to be something which
is not analyzable further. Even from this perspective, excluding one-place predicates to
introduce two-place predicates does not make much sense either. Therefore, I agree with
what Russell says (against Loux’s observations in Substance and attribute, cit., 45–46) in
Ch. 9 of Problems of Philosophy. The seemingly promising route followed by some con-
temporary nominalists is to insist on the differences in semantic behavior between prop-
er nouns and predicates. Perhaps the latter position can provide a starting point for an
interpretation of the quid aliquid which lends more dignity to what John writes about the
objects denoted by nouns.
John of Salisbury on Universals 297
The unclear points in the theses above induce John to grapple with new
solutions. However, we cannot rest confident that we have found a textual
match or a new standpoint, e.g. the Boethian theory of similitudo or a different
position on the relation of similarity. John continues by producing other pos-
sible eligible universal entities: the genera and species as qualified quids, ‘ex
sententia Aristotelis,’ are first presented as figmenta rationis, then as monstra
and finally ‘secundum novam translationem’ as chatter (cicadationes).46 John,
however, seems to take none of these seriously:
Genera and species are exemplars of individual things. But this applies
more to the understanding of science (if Aristotle is right) than of es-
sence, and this monstrous speculation about fictional things – if so it
may be called – goes on until one is exposed to the fresh air of particular
things.47
This monstrosity, however, has already played a role in the discussion at hand.
Ultimately, we see that John has tried to consolidate a position of strict nomi-
nalism (in the modern sense) on an island surrounded by more or less mon-
strous things.48 In order to visualize what surrounds the fort of ‘bona singula
creata’ it is useful to show the path we have followed in a diagram:
46 Some literary suggestions about the terminology of ‘monsters’ can be found in T. Goeglein,
‘The Problem of Monsters and Universals in The Owl and the Nightingale and John of
Salisbury’s Metalogicon’, Journal of English and Germanic philology, 94 (1995), 190–206.
47 Metal. 97, 431–435: ‘Sunt itaque genera et species exemplaria singulorum; sed hoc quidem
magis ad rationem doctrinae (si Aristoteles verum est) quam ad causam essentiae; pro-
cedit et haec monstruosa (ut licentius loquar) figmentorum speculatio usque ad ventila-
tionem singularium.’
48 John’s theoretical position (as well as that of Abelard) has been interpreted in many dif-
ferent ways. B. Hendley (‘John of Salisbury and the Problem of Universals’, Journal of the
History of Philosophy, 8 (1970), 289–302) lists the following proposals: the solution of com-
mon sense (Shapiro), basic Aristotelianism (Webb); Platonic nominalism (Peirce); no
philosophically interesting solution (Prantl); a philosophically interesting original solu-
tion (Denis); moderate, confused realism (de Wulf); or a position in some way linked to
that of Quine on ontology (D.P. Henry). The author of the list is tempted to add the new
tag: ‘form of conceptualism with Lockean nuances’.
298 Chapter 10
concepts
fantasie
exemplars similitudi-
nes
cigadatio- exemplars
nes in the
mind
figmenta cogitabilia
imagines monstra
conformi- types
tates
res in
relations some
sense
qualia qualia
quid
Primary Sources
Abelard, Logica Ingredientibus (= Log. Ing.), ed. B. Geyer, Aschendorff, Münster 1919–27
(BGPhM, XXI, Heft 1–3).
Abelard, Dialectica (= Dial.), ed. de Rijk, Van Gorcum, Assen 1970.
Abelard, Logica ‘Nostrorum petitioni sociorum’ (= Log. Nostr.), ed B. Geyer, Aschendorff,
Münster 1933 (BGPhM, XXI, Heft 4).
Abelard, Historia calamitatum, Ed. J. Monfrin, Vrin, Paris 1967.3
Abelard, Introductiones Dialecticae, ed. M. Dal Pra, in Pietro Abelardo, Scritti di Logica,
La Nuova Italia, Firenze 19692.
Abelard, Editio super Porphyrium (= Ed. Sup. Por.), ed. M. Dal Pra, in Pietro Abelardo,
Scritti di Logica.
Abelard, Glossae in Categorias (= Gloss. In Cat.), ed. M. Dal Pra, in Pietro Abelardo, Scritti
di Logica.
Abelard, Super Topica Glossae (= SupTop.), ed. M. Dal Pra, in Pietro Abelardo, Scritti di
Logica Theologia Christiana, ed. E.M. Buytaert, (CCCM), Turnhout, Brepols, 1969.
Adam Parvipontanus, Ars disserendi, in L. Minio-Paluello, Twelfth century logic. Texts
and studies 1. Adam Balsamiensis Parvipontani, Ars disserendi (Dialectica Alexandri),
Edizioni di storia e letteratura, Roma 1956.
St. Anselm, Opera Omnia, ed F.S. Schmitt, Fromman Verlag, Stuugart – Bad Cannstatt
1984.
Aristotle, Prior Analytics (= An. Pr.), in W.D. Ross (ed.), Aristotle’s Prior and Posterior
Analytics, Clarendon Press, Oxford 1949.
Aristotle, Categories, De l’interprétation, traduction nouvelle et notes par J. Tricot, Paris, Vrin,
1966.
Aristotle, Le categorie, introduzione, traduzione e commento di Marcello Zanatta, Milano,
Rizzoli, 1989.
Aristotle, Aristotle’s Categories and De interpretatione, translated with notes by J.L. Ackrill,
Oxford, Clarendon Press, 1963.
Aristotle, Le Categorie, a cura di D. Pesce, Liviana, Padova 1966.
Bernard of Chartres, Glosae super Platonem, Pontifical Institute of Mediaeval Studies,
Toronto 1991.
Boethius, In Categorias Aristotelis (= In Cat.), in PL 64.
Boethius, In Isagogen Porphyrii Commenta (In Isag.), ed. S. Brandt, Tempsky – Freytag,
Vienna – Lipsia 1906.
Boethius, Commentarii in librum Aristotelis Peri hermeneias I–II, ed. C. Meiser, 2 voll.,
Lipsiae, 1877–1880 (= In De Int.).
300 Bibliography
28, 1990), 93–94; ed. Romano (in Una soluzione originale della questione degli uni-
versali nel XII secolo, Aracne, Roma 2007).
William of Sherwood, Introductiones in Logicam, ed. M. Grabmann, Sitzungsberichte
der Bayerischen Akademie der Wissenschaften. Philosophisc – historische Abteilung,
1937, Heft 10, Munchen 1937.
Secondary Sources
Amini, M., ‘Logical machines: Peirce on psychologism’, Disputatio, II-24 (2008), 335–348.
Amstrong, D.M., Universals. An Opinionated Introduction, Westview Press, Boulder –
London 1989.
Angelelli, I., Studies on Gottlob Frege and traditional philosophy, Reidel 1965.
Apostle, H.G., Aristotle’s Philosophy of Mathematics, The University of Chicago Press,
Chicago 1952.
Arlig, A.W., (A Study in Early Medieval Mereology: Boethius, Abelard and Pseudo-Joscelin.
Dissertation for the Degree, The Ohio State University, 2005.
Arlig, A.W., ‘Early medieval solutions to some mereological puzzles: the content and
unity of the Generibus et Speciebus’, in Arts du language et théologie aux confins
des Xie / XIIe siècles, ed. I. Rosier-Catach, Brepols, Turnhoult 2011, 485–508.
Bacon, J., ‘Tropes’ in E.N. Zalta, The Stanford Encyclopedia of Philosophy, 2008.
Beierwaltes, W., ‘The revaluation of John Scottus Eriugena in German Idealism’, in
J. O’Meara – L. Bieler (eds) The Mind of Eriugena, Irish University Press Dublin 1973,
pp. 190–199.
Beierwaltes, W., Eriugena. Grundzüge seines Denkens, Klostermann, Frankfurt am Main
1994.
Bobzien, S., ‘Stoic logic’ in K. Algra – J. Barnes – J. Mansfeld – M. schofield (edd.), The
Cambridge History of Hellenistic Philosophy, Cambridge University Press, Cambridge
1999, 92–157.
Bottin, F., ‘Quelques discussions sur la transitivité de la predication dans l’ècole
d’Alberic du Mont’, in J. Jolivet – A. De Libera (edd.), Gilbert de Poitiers et ses contem-
porains. Aux origines de la Logica Modernorum, Bibliopolis, Napoli 1987.
Brower, J.E., and Guilfoy, K., (edd), The Cambridge Companion to Abelard, Cambridge
UP, Cambridge 2004.
Burnett, Ch., ‘John of Salisbury and Aristotle’, Didascalica 2 (1996), 19–32, disponi-
bile all’indirizzo: www.scribd.com/doc/26824811/John-of-Salisbury-and-Aristotle
-Charles-Burnett-Didascalica-2-1996.
Carabine, D., The Unknown God: Negative Theology in the Platonic Tradition: Plato to
Eriugena. Louvain – Grand Rapids, Peeters-Eerdmans 1995.
Clanchy, M.T., Abelard: A Medieval Life, Blackwell, Oxford 1997.
302 Bibliography
Code, A., ‘Aristotle: essence and accident’, in R. Grandy – R. Warner (edd.) Philosophical
Grounds of Rationality: Intentions, Categories, Ends, Oxford University Press, Oxford
1983, 411–439.
Cohen, Marc S., ‘Grice and Code on IZZing and HAZZing’: scaricabile dal sito web:
http://faculty.washington.edu/smcohen/433/GriceCode.pdf, 2008.
Corazzon, R., Abelard bibliography: www.ontology.co/biblio/abelard-biblio-one.htm,
www.ontology.co/biblio/abelard-biblio-two.htm).
Courcelle, P., ‘Boèce et l’école d’Alexandre’, Mélanges dìarcheologie et d’histoire de
l’Ecole francaise de Rome, 1935, 185–223.
Courcelle, P., Les lettres grecques en Occident. de Macrobe à Cassiodore, Paris 1948.
Cristiani, M., ‘Nature-essence et nature-language. Notes sur l’emploi du terme “natu-
ra” dans le Periphyseon de Jean Scot Erigène’. In J. Beckmann (ed.), Sprache und
Erkenntnis im Mittelalter. Akten des VI. internationalen Kongresses für mittelalterli-
che Philosophie der Société internationale pour l’étude de la philosophie médiévale,
29. August-3. September 1977 im Bonn. Walter de Gruyter, Berlin 1981. pp. 707–717.
Cross, R., ‘Form and universal in Boethius’, British Journal for the History of Philosophy,
20, 2012, 439–58.
de Libera, A., La querelle des universaux de Platon à la fin du Moyen Age, Éditions du
Seuil, Paris 1996.
de Libera, A., L’art des généralités, Aubier, Paris, 1999.
de Rijk, L.M., Logica Modernorum: A Contribution to the History of Early Terminist Logic.
I. On the Twelfth Century Theories of Fallacy, Assen, Van Gorcum, 1962.
de Rijk, L.M., Logica Modernorum II-1, The Origin and Early Development of the Theory of
Supposition, Assen, Van Gorcum, 1967.
de Rijk, L.M., ‘The Semantical Impact of Abelard’s Solution of the Problem of Uni-
versals’, in R. Thomas – L.M. de Rijk (edd.) Petrus Abaelardus (1079–1142) Person
Werk und Wirkung (Trierer Theologische Studien 38), Paulinus-Verlag, Trier 1980,
139–151.
de Rijk, L.M., ‘Peter Abaelard’s semantics and his doctrine of being’, Vivarium 24 (1986),
85–127.
de Rijk, L.M., ‘Gilbert de Poitiers, ses vues sémantiques et métaphysiques’, in J. Jolivet –
A. de Libera (edd.), Gilbert de Poitiers et ses contemporains. Aux origines de la Logica
Modernorum, Bibliopolis, Napoli 1987, 147–171.
de Rijk, L.M., ‘Semantics and metaphysics in Gilbert of Poitiers. A chapter of twelfth
century Platonism’, Vivarium XXVI, 2 (1988), 73–112.
de Rijk, L.M., Semantics and Ontology, Vol. 1: General Introduction. The Works on Logic,
Brill, Leiden 2002.
Devitt, M., and Sterelny, K., Language and Reality: An Introduction to the Philosophy of
Language, Blackwell, Oxford 1999.
Bibliography 303
Dijs, J., ‘Two Anonymous 12th-Century Tracts on Universals’, Vivarium, 28, 1990,
pp. 85–113.
Dutton, P.E., ‘The uncovering of the Glosae Super Platonem of Bernard of Chartres’,
Mediaeval Studies, 1984, 192–221.
Dutton, P.E., ‘Introduction’ in: Bernard of Chartres, Glosae super Platonem, Pontifical
Institute of Mediaeval Studies, Toronto 1991.
Ebbesen, S., ‘Medieval latin glosses and commentaries on aristotelian logical texts of
the twelfth and thirteenth centuries’, in C. Burnett (ed.), Glosses and Commentaries
on Aristotelian Logical Texts. The Syriac, Arabic and Medieval Latin Traditions, The
Warburg Institute, London 1993, 129–147.
Ebbesen, S., ‘The Aristotelian commentator’ in: John Marenbon, ed., The Cambridge
Companion to Boethius, Cambridge University Press, pp. 33–55, Cambridge 2009.
Erismann, C., ‘Processio id est multiplicatio. L’influence latine de l’ontologie de Porphyre:
le cas de Jean Scot Érigène’, Revue des Sciences Philosophiques et Théologiques 88
(2004), 401–460.
Erismann, ‘Alain de Lille la metaphysique érigénienne et la pluralité des formes’
(in J. Solère – A. Vasiliu – A. Galonnier (eds), Alain de Lille, le Docteur universel.
Philosophie, théologie et littérature au XIIe siècle. Actes du XIe Colloque interna-
tional de la Société internationale pour l’Étude de la Philosophie médiévale, Paris,
23–25 octobre 2003, Brepols, Turnhout 2005. pp. 19–46.
Erismann, La gènese du réalisme ontologique durant le haut Moyen age, Lausanne 2006.
Erismann, ‘The logic of being: Eriugena’s dialectical ontology’, Vivarium 45 (2007),
203–218.
Erismann, ‘Explaning exat resemblance. Gilbert of Poitiers’s conformitas theory recon-
sidered’ in R. Pasnau (ed.), Oxford Studies in Medieval Philosophy, vol 2, OUP, Oxford
2014, 1–24.
Forrest, P., ‘The identity of Indiscernibles’, in E.N. Zalta (ed.), The Stanford Enciclopedia
of Philosophy, 1996, rev. 2010.
Fournier, M., ‘Eriugena’s five modes (Periphyseon 443A–446A)’, The Heythrop Journal
2009, 581–589.
Gersh, S., From Iamblichus to Eriugena, Brill, Leiden 1978.
Gersh, S., ‘Eriugena’s fourfold contemplation. Idealism and arithmetic’, in S. Gersh –
D. Moran (edd.), Eriugena, Berkeley and the Idealist Tradition, University of Notre
Dame Press, Notre Dame 2006, 151–167.
Goeglein, T., ‘The problem of monsters and universals in The Owl and the Nightingale
and John of Salisbury Metalogicon’, Journal of English and Germanic Philology, 94
(1995), 190–206.
Gracia, J.J.E., Introduction to the Problem of Individuation in the Early Middle Ages,
Philosophia Verlag, München – Wien 1984.
304 Bibliography
Grellard, C., and Lchaud, F., (ed.), A Companion to John of Salisbury, Brill, Leiden 2014.
Grice, H.P., ‘Aristotle on the multiplicity of being’, Pacific Philosophical Quarterly, 69
(1988).
Guha, S., ‘Predication and exemplification in Aristotle’s categories’, web: www.saikat-
guha.com/metaphysics/predication-and-exemplification-in-aristotles-categories.
Guilfoy, K., ‘John of Salisbury’, in E.N. Zalta (ed.), The Stanford Encyclopedia of
Philosophy, 2005.
Hauréau, B., Notices et extraits de quelques manuscripts latins de la Bibliothèque
Nationale, II, Paris 1891, 298–333.
Hendley, B., ‘John of Salisbury and the problem of universals’, Journal of the History of
Philosophy, 8 (1970), 289–302.
Hülsen, R., ‘Understanding the semantics of ‘relativa grammaticalia’. Some medieval
logicians on anaphoric pronouns’, in K. Von Heusinger – U. Egli, Reference and
Anaphoric Relations, Kluwer, Dordrecht 2000, 31–46.
Inwood, B., (ed.), Oxford Studies in Ancient Philosophy, vol XL, OUP, Oxford 2011.
Iwakuma, Y. ‘‘Vocales’ or early nominalists’, Traditio, XLVII, 1992.
Iwakuma, Y., ‘Twelfth-Century Nominales. The Posthumous School of Peter Abelard’, in
Vivarium 1992 (XXX, 1).
Iwakuma, Y., ‘Influence’, in J.E. Brower and K. Guilfoy, The Cambridge Companion to
Abelard, Cambridge UP, Cambridge 2004, 305–335.
Jacobi, K., ‘Abelard and Frege: the Semantics of Words and Propositions’, in Atti del con-
vegno internazionale di storia della logica, S. Gimignano, 4–8 dicembre 1982, Clueb,
Bologna 1983, 81–96.
Jacobi, K., ‘Logic (ii): The later twelfth century’, in P. Dronke (ed.), A History of Twelfth-
Century Western Philosophy, Cambridge UP, Cambridge 1988, pp. 227–235.
Jacobi, K. ‘Philosophy of language’, in J.E. Brower – K. Guilfoy (edd), The Cambridge
Companion to Abelard, Cambridge UP 2004, 126–157.
Jaeger, C.S., ‘Irony and role-playing in John of Salisbury and Becket circle’, in M. Aurell
(ed.), Culture politique des Plantagenêt, 1154–1224 (Actes du colloque tenu à Poitiers
du 2 au 5 mai 2002), CESCM, Poitiers, 2003, 319–331.
Jeauneau, E., ‘Le renoveau érigénien du XIIe siècle’, in W. Beierwaltes (ed.), Eriugena
redivivus. Zur Wirkungsgeschichte seines Denkens in Mittelalter und im Übergang
zur Neuzeit (Vorträge des V. Internationalen Eriugena-Colloquiums Werner-
Reiners-Stiftung Bad Homburg 26.-30. August 1985), Carl Winter Universitätsverlag,
Heideberg 1987, 26–46.
Jolivet, J., Arts du language et théologie chez Abelard, Paris, Vrin, 1982.
Jolivet, J., Aspects de la pensée médiévale: Abélard. Doctrines du language, Vrin, Paris 1987.
Kavanagh, C., ‘The influence of Maximus the Confessor on Eriugena’s treatment
of Aristotle’s Categories’, American Catholic Philosophical Quarterly, 79 (2005),
567–596.
Bibliography 305
Marenbon, J., ‘Medieval Latin Commentaries and Glosses on Aristotelian Logical Texts
Before c. 1150 AD’, in J. Marenbon, Aristotelian Logic, Platonism and the Context of
Early Medieval Philosophy in the West, Aldershot, Ashgate 2000, item II.
Marenbon, J., Boethius, Oxford U.P., Oxford 2003.
Marenbon, J., ‘Life, milieu, and intellectual contexts’, in J.E. Brower – K. Guilfoy (edd),
The Cambridge Companion to Abelard, Cambridge UP 2004, 13–44.
Marenbon, J., ‘Gilbert of Poitiers’, Cap. 35 di: J.E. Gracia – T.B. Noone, A Companion to
Philosophy in the Middle Ages, Blackwell, Malden 2005.
Marenbon, J., ‘Logic at the turn of the twelfth century’, in D..M. Gabbay – J. Wood (eds.),
Handbook of the History of Logic, vol 2 Medieval and Renaissance Logic, Elsevier,
Amsterdam 2008, 65–81.
Marenbon, J., Abelard in Four Dimensions. A Twelfth-Century Philosopher in his Context and
Ours, University of Notre Dame Press, Notre Dame 2013.
McEvoy, J., ‘Metaphors of light and metaphysics of light in Eriugena’, in W. Beierwaltes
(ed.), Begriff und Metapher. Sprachform des Denkens bei Eriugena, Carl Winter,
Heidelberg 1990; pp. 149–167.
McGarry, D., The Metalogicon of John of Salisbury: A Twelfth Century Defense of the
Verbal and Logical Arts of the Trivium, University of California Press, Berkeley, 1955
(rist. Paul Dry Books, Philadelphia, 2009).
McGinn, B. and Otten, W., Eriugena: East and West (Papers of the Eight International
Colloquium of the Society for the Promotion of Eriugenian Studies, Chicago and
Notre Dame 18–20 October 1991) University of Notre Dame Press, Notre Dame
1994.
McLaughlin, B., and Bennett, K., ‘Supervenience’, in E.N. Zalta (ed.), The Stanford
Encyclopedia of Philosophy, rev. 2011.
Mews, C.J., ‘On dating the works of Peter Abelard’, Archives de l’histoire doctrinale et
littéraire de moyen age, 52 (1985), 73–134.
Mignucci, M., ‘Puzzles about identity. Aristotle and his greek commentators’, in
P. Moreaux – J. Wiesner (edd.), Aristoteles Werk und Wirkung. Erster band. Aristoteles
und seine schule, De Gruyter, Berlin – New York 1985.
Mill, Stuart J., A System of Logic Ratiocinative and Inductive, J.W. Parker, London 1863,
Vol. 1–2.
Moran, D., The Philosophyof Jhon Scottus Eriugena. A Study of Idealism in the Middle
Ages, Cambridge University Press, Cambridge 1989 (rist. 1990).
Moran, D., ‘Idealism in medieval philosophy: the case of Johannes Scottus Eriugena’,
Medieval Philosophy and Theology, 8 (1999), 53–82.
Muller, I., ‘Aristotle on Geometrical Objects’, Archiv für Geschichte der Philosophie, 52
(1970), 156–171.
Hochschild Paige, E., ‘Ousia in the Categoriae decem and the Periphyseon of John
Scottus Eriugena’, in M. Treschow – W. Otten – W. Hannam (edd.), Divine Creation
Bibliography 307
in Ancient, Medieval, and Early Modern Thought. Essays Presented to the Rev’d Doctor
Robert D. Crouse, Brill, Leiden 2007. pp. 213–222.
Obertello, L., Severino Boezio, Accademia ligure di scienze e lettere. Genova 1974.
O’Meara, D., ‘The concept of natura in John Scottus Eriugena (De divisione naturae
Book I)’, Vivarium XIX, 2 (1981), 126–145.
O’Meara, D., ‘The metaphysical use of mathematical concepts in Eriugena’, in
W. Beierwaltes (ed.), Begriff und Metapher: Sprachform des Denkens bei Eriugena,
Carl Winter, Heidelberg 1990, 142–148; rist. In D. O’Meara, The Structure of Being and
the Search for the God, Ashgate, Aldershot-Brookfield 1998, cap. XXIII.
D’Onofrio, G., ‘A proposito del ‘magnificus Boetius’: un’indagine sulla presenza degli
Opuscola sacra e della Consolatio nell’opera eriugeniana’, in W. Beierwaltes (ed.),
Eriugena. Studien zu seinen Quellen, Carl Winter, Heidelberg 1980; pp. 189–200.
D’Onofrio, G., ‘Agli inizi della diffusione della Consolatio e degli Opuscula sacra nella
scuola tardo-carolingia: Giovanni Scoto e Remigio di Auxerre’, in L. Obertello (ed.),
Atti del Congresso Internazionale di Studi boeziani (Pavia, 5–8 ottobre 1980), Herder,
Roma 1981. pp. 343–354.
D’Onofrio, G., ‘Dialectic and theology: Boethius’ Opuscula sacra and their early medi-
eval readers’, Studi Medievali 27 (1986), 45–67.
Otten, W., ‘The shadow of the divine: negative theology and negative anthropology
in Augustine, Pseudo-Dionysius and Eriugena’, The Heythrop Journal, 40 (1999),
438–455.
Paolucci, P., ‘Boezio traduttore di Nicomaco nel De Institutione Arithmetica’, Atheneum
93 (2005), 227–241.
Parret, H., History of Linguistic Thought and Contemporary Linguistics, de Gruyter,
Berlin 1976.
Petroff, V., ‘Eriugena on the spiritual body’, American Catholic Philosophical Quarterly,
79 (2005), 597–610.
Pinzani, R., La logica di Boezio, Angeli, Roma 2003.
Pinzani, R., The Logical Grammar of Abelard, Kluwer, Dordrecht 2003.
Pinzani, R., ‘L’origine del linguaggio nei commenti logici medievali’, in F. Amerini –
R. Messori (edd.), Sulle origini del linguaggio. Immaginazione, espressione, simbolo,
ETS, Pisa 2012, 79–91.
Priest, G. and Tanaka, K., ‘A brief history of ex contradictione quodlibet’: Cap. 3 di:
‘Paraconsistent logic’, in E.N. Zalta (ed.), The Stanford Encyclopedia of Philosophy,
2009.
Priest, G. and Tanaka, K., ‘Paraconsistent logic’, in D.M. Gabbay – F. Guenthner (edd.),
Handbook of Philosophical Logic, Vol. 6, Kluwer 2002, 287–393.
Priest, G. and Tanaka, K., ‘Paraconsistency and dialetheism’, in D.M. Gabbay – J. Woods
(edd.), Handbook of the History of Logic, Vol. 8: The Many Valued and Non-Monotonic
Turn in Logic, North Holland (Elsevier), Amsterdam – Oxford 2007, 129–204.
308 Bibliography
Abelard 20, 87–88, 91–114, 119, 129, 136–145, Collection(s) 88–91, 104, 107, 109, 170–172,
151–153, 156, 158, 163–166, 169, 171–172, 175, 178–179, 187, 189–192, 194, 199–200,
175, 177–178, 180, 184–188, 191n45, 192, 205–206, 212, 228, 261, 267, 269, 290
194, 198–234, 247, 269–272, 274–275, Common cause of imposition 109, 185, 198,
277, 279, 281, 286, 289–293, 295 212–213, 234, 269, 292
Act and potency 96–98, 153–154, 174 Concept(s) 19–20, 29, 38–40, 48–52, 107, 111,
Adam Parvipontanus 272 148, 174, 193, 210–211, 214–215, 224–26,
Alberic of Reims 271–272 274–276, 282, 285–287, 292, 296, 298
Albino 50n57 Condition (definition) of truth 30–33,
Alexander 49, 147 188–190, 209, 231, 268, 287
Amerini, F. 211n20 Conformity 226, 254, 258–261, 267, 279–280,
Amini, M. 272n6 296
St. Anselm 115, 213, 227 Corazzon, R. 228n
Aristotle 5–6, 8–12, 14, 18–20, 22–23, 25, 27, Courcelle, P. 51n57
29–32, 39, 45, 47, 66, 74, 79–80, 85, 89, Cousin, V. 170n1
96, 98, 107, 113–115, 123–125, 136–138, Cristiani, M. 57n6
143–144, 155, 177, 199, 214n, 215, 227, 241, Cross, R. 51n57
246–247, 273–275, 279, 281–282, 293
Arlig, A. 170 Dal Pra, M. 136n26
Armstrong, A.H. 57n8 Definite descriptions 27, 31, 229
Armstrong, D.M. 20–21 Denis, L. 297n48
Asher, N. 17n19 Denotation (nominatio) 23–31, 72, 75, 77, 87,
St. Augustine 55, 67, 285n 111, 115–116, 141, 145, 163–164, 184,
187–188, 193, 211–212, 223, 225, 227–229,
Barth, E.M. 17 256, 267, 278–279, 289, 291–294
Beierwaltes, W. 86n70 De Rijk, L.M. 152n15, 173n9, 185n31, 198n1,
Bernard of Chartres 277–279, 285n 223n38, 235n, 240
Bobzien, S. 227n45 Determinative meaning (determinatio)
Boethius 11–13, 19–20, 22–54, 57n6, 65, 25–31, 108, 141–142 293
70–71, 75–76, 79–81, 83–84, 88–89, 91, Devitt, M. 224n40
95–98, 106, 108–111, 114–115, 120, 129, De Wulf, M. 297n48
146–150, 154, 157, 162, 165–166, 177, 180, Difference (-diversification) 157, 158–169,
185n31, 195–196, 199, 215, 217–218, 172, 174–178, 181, 183, 189, 200–203, 216
221–222, 224–225, 230, 235, 238–239, Dijs, J. 146n1, 151, 158n22, 159n24, 164n, 167n
241, 243–250, 256, 258, 260–261, Diogener Laertius 227n45
268–269, 276–279, 281–285, 296 Dionysius the Areopagite 58n116, 83n63
Bottin, F. 13n15 Dividual (dividuum) 257, 259–261, 264
Brower, J.E. 151n12, 171n3, 198n1 Donatus 185n31
Burnett, Ch. 271n1 D’Onofrio, G. 65n23
Dutton, P.E. 277n, 278n15
Chalcidius 285n
Carabine, D. 58n11 Ebbesen, S. 136n27
Clanchy, M.T. 198n1 Erismann, C. 56n4, 59n13, 80n56, 254n44,
Code, A. 13n14 267n64
310 Index
Essence 6, 13–16, 19, 46, 48, 58–61, 64–74, 78, Inherence 30–31, 44, 108, 110–113, 176,
82, 85–86, 91, 94–95, 101, 104–107, 151, 188–191, 218, 231–233, 237, 242, 252, 258,
179–183, 187–197, 200, 202–203, 216, 222, 265–267, 279
228, 235–236, 255, 266, 277, 280 intensional (contexts-expressions-
Example-exemplar 284, 277–279, 284–285, interpretations) 14, 121 133, 227, 260, 289,
298 290
Iwakuma, Y. 114n2
Fine, K. 17n19
Forms 15, 19, 27, 29, 34, 39, 40, 43–44, 49, 52, Jaeger, C.S. 279n18
54, 65, 71–77, 89–96, 108, 110, 157, 179, Jakobi, K. 198n1, 230n50, 233n, 271n1
182, 189, 191, 197, 200, 202–203, 205, 216, Jeauneau, E. 56n4
225, 243, 251–252, 261–264, 295–296 John of Salisbury 106–108, 146n2, 171, 245,
Forrest, P. 159n25 254, 271–298
Fournier, M. 60n15 Jolivet, J. 13n15, 185n31, 212n23
Frege, G. 230n50 Joscelin of Soissons 91, 94, 105,107, 171–197,
254n44, 255, 261, 280
Garland the Computist 114–139, 141, 145, 159,
177–178 Keats-Roha, K.S.B. 271n1
Gersh, S. 57n8 King, P. 48n51, 170n1, 198n, 228n, 230n50
Gilbert of Poitiers 21–22, 70, 88, 91, 235–270, Kneale, W.C. 9–10, 272n6
273, 279–280 Kneepkens, C.H. 271n1, 288n, 290n37
Goeglein, T. 297n46 Knuuttila, S. 97n14,
Goodman, N. 115
Gregory of Nyssa 55, 59n13, 67, 74 Lchaud, F. 271n1
Grellard, G. 271n1 Leibniz 15, 159, 260
Grice, H.P. 13n14 Lejeune, F. 271n1
Guilfoy, K. 151n12, 171n3, 198n1, 271n1 Libera, A. de 6n3, 13n15, 50n57, 51n60
Locke, John 69,
Hall, G.B. 171n3 Loux, M.J. 5–6, 69n36, 296n
Haureau, B. 146n1, 147, 148n3, 149n6 Luscombe, D.E. 185n31
Heidegger, M. 56
Hendley, B. 297n48 Maioli, B. 253n44
Henry, D.P. 297n48 Maloney, J.C. 228n
Heraclitus 278 Manner (maneria) 120, 281
Hochschild Paige, E. 65n22 Marenbon, J. 50n57, 56n3, 65n22, 80n56,
Hulsen, R. 288n 81n58, 115, 136n, 198n1, 235n, 239,
240n19, 253
Ideas 107, 215, 246, 268, 277–79, 285 Marius Victorinus 33
Identity 80, 99–105 109, 146–149 157–159, Magister R. 107
163–164, 167–69, 176–177, 188–192, Magister W. 164
202–203, 207, 215, 217–219, 255–256, Master G. 171
259–260, 268, 287, 296 Matter-form 62, 66–68, 70–75, 77, 92, 95,
Id quod-esse (quo est) 70, 235, 237–240, 243, 102, 107, 138, 141, 152–153, 163, 178–179,
265 195–196, 203, 216, 232, 241–243, 250,
Impositive meaning (- impositio) 125, 130, 294
164, 184, 186–187, 208, 211–213, 221–222, McEvoy, J. 83n63
293 McGarry, D.D. 277n, 290n35
Index 311