Unidad 2. The Concept of Morpheme
Unidad 2. The Concept of Morpheme
Unidad 2. The Concept of Morpheme
1. Morpheme.
‐ Morpheme is the minimal grammatical unit with form and meaning, the minimal meaningful
unit of grammatical analysis.
‐ Bloomfield, 1933: “Morpheme is a linguistic form that bears no partial phonetic‐ semantic
resemblance to any other form”. There’s no partial resemblance to any other form= within a
morpheme we cannot find parts which are similar to other parts either semantically or
phonetically.
For example, dancing can’t be a morpheme-> danc‐ = danc‐ in dances -> ‐ing = ‐ing in singing
(dancing has partial resemblance with other forms, that is, parts of it are similar to other
forms. Therefore, it is NOT a morpheme. It contains 2 morphemes but is not a morpheme as a
whole). The whole unit is not a morpheme (dancing) but inside it has the morpheme -ing.
For example, un‐happy -> ‘not’- unhappy has two elements: happy and un. They are bot
morphemes. However, about the idea of a morpheme as a minimum unit, there are cases in
which we can recognise constant realizations but without the same meaning (pre‐fer, in‐fer,
de‐fer, con‐fer, trans‐fer, re‐fer-> the meaning of fer- was present in the word in their origin
but it is not any longer).
Possible analyses are that there are two morphs/ morphemes (one meaningless, exceptional
type of morpheme without meaning) or that it is a monomorphemic word, or that there is a
morph/ morpheme and another element called formative (a more general concept), used to
form words but that has no meaning.
Consider also:
b) Inflectional formants: ‐s (read‐s). They are inflectional affixes with a more grammatical
and abstract meaning.
‐ Lyons (1968): “The components, or distributional factors, of the words are morphemes”,
different syntactic positions the word can occupy on a sentence depends on the morphemes it
is made of. He talks about the distribution of words; distribution is the different syntactic
arrangement where they can be used.
And
b) Distributionally different (qualify different classes of nouns, tall mountain, tall
building, tall man/*smart man, but smart building? , smart mountain?).
Morphemes are abstract units (not real units) that derive from the linguist’s analysis. We cant
represent them unless writing (orthographic representation) it or speaking (phonological
realization).
2. Morph.
Word form – Lexeme (all lexemes need a basic word form, used in real syntactic structures)
The ideal situation would be the correspondence of one morph to one morpheme, like in
Swahili. However:
‐ The same form (morph) may be used to represent different morphemes. (The opposite
case as the one before). Ex: homophones: words that have different meaning and spelling but
the same pronunciation, ej) ate /eight.
‐ The same form (morph) may represent several morphemes at the same time. Rosa,
rosae, rosas
3.1. Allomorph.
Morphs don’t belong to a morpheme, it contains. The teacher doesn’t like this definition.
a) After voiced segments other than sibilants: /z/ assimilation of voiced phonemes (dogs).
- Genitive Morpheme
a) After voiced segments except sibilants: /z/ ej) Mick’s
b) After voiceless consonatns except sibilants /s/ (assimilation) Jane’s
c) After sibilants (/s,z, ʃ, tʃ, ʒ, dʒ/): /lz/ (dissimilation) Liz’s
‐ Past tense morpheme:
a) After voiced segments other than /d/: /d/ assimilation of voiced phonemes (warned).
c) After alveolar plosives (/t,d/): /id/ dissimilation, vocalic support might be needed
(added).
‐ Indefinite article:
c) In isolation: /eɪ/
‐ Definite article:
c) In isolation: /ðɪ/
‐ Complementary distribution: one variant (‐al) is exclusively found in one environment, while
the other variant (‐ar) is exclusively found in a different environment.
MORPHOLOGICALLY CONDITIONED
‐ The number of syllables determine the form of the past tense. The internal change in
the root concord by the presence of the past morpheme: sweep, past
Because….
LEXICALLY CONDITIONED
Homophones: The same phonological form (morph) is used to represent different morphemes,
but written English uses a different orthographic form to represent each morpheme. For
example, /sait/ (sight, site, cite)- the orthography shows that they are different morphemes,
the problem arises when we hear them.
Syncretism: the same form is used to represent different morphemes, but this time the
orthographic representations are identical, although they represent different grammatical
words. For example, the past and past participle of regular verbs (continued – has continued),
the singular and plural forms of the second person pronoun “you”. No different orthographic
and phonological form. The morphosyntax properties are different. Cut and car are more
considered case of conversion (like banco in ES).
Hockett: A single morph that represents different morphemes at the same time. There is no
possible division into different segments representing each of the meanings expressed.
For example, ‐s {present tense} + {sg} + {3rd person}, was {BE} + {sg} + {past}.
Bloomfield called it zero allomorph or zero morpheme. They are morphemes that are present
but are not given any phonetic manifestation. For example, sheep (plural, inflection), hit (past,
inflection- no representation of the past meaning in the word/ could and must be in the
sentence), drive (noun, conversion, derivation- Where is the mark of this derivation process?
There isn’t. ).
3.5. Empty morph. > opposite case of Portmanteau morph
Why call them morphs? -> Though they have no meaning, they have a function contributing to
the construction of words (Bauer).
Alternative analyses: maybe the base has two allomorphs (base allomorphy) (Platon ‐ Plato) or
the suffixes are allomorphs (suffix allomorphy) (‐nic and ‐ic).
Bauer proposes the term ‘formative’, bc linguistics don’t like the idea of morpheme as a
meaningless unit. These are cases of unit contributing to the meaning of a word but without a
meaning.
3.6. Formative.
Bauer: An overarching category that includes both morphemes and elements contributing to
the construction of words whose semantic unity or function is obscure or dubious. That is to
say, an element that recurs in the morphological analysis of word‐ forms, a surplus word‐
building element.
Morphemes are not realized by affixes but by means of processes, an instruction to replace
one item by another in the phonological make‐up of the base. Cases:
‐ Vowel alternation between verbs and corresponding nouns: shoot‐shot, sing‐ song.
All the previous cases weaken the idea that morpheme is a unit, a sign, an element, an entity
with a physical representation. In all these cases the morpheme is a process.
3.7. Suppletion.
Notice that in all these cases suppletion concerns the base of word, that is, the comparative
form of good is no gooder, but better. Therefore, there is not suppletion in the suffix. The
comparative form is still regular; it is the root of the word which change.
Ex: the plural mark children/ dogs -> different endings as a mark pf plural.
en in oxen (Bloomfield)
Alternative analyses:
Problem with suppletion -> It is not always easy to determine where the boundaries of
suppletion lie, that is, to decide whether we have lack of morphological relatedness of the
same base / affix (suppletion) or multiple bases / affixes (not different realizations but different
morphemes).
CONCLUSION
‐ The notion of morpheme is not without problems: there are problematic cases where the
number of formal elements and the number of semantic elements does not seem to match
neatly.
‐ Anderson: There never was, and certainly is not today, a single coherent view held by all
theorists of what the morpheme is. The very fact that scholars cannot agree on the nature of
the morpheme is a weakness.
‐ The real problem with the morpheme I that it is not maximally economical. We tend
to use notions, mechanisms that accounts for as many cases as possible. We have to still
propose other notions.
Some scholars, like Anderson argues that the notion of morpheme is inherently flawed (not
good in itself). They operate with the word‐form as the basic unit, within which they identify
morphs, but without the link between morph and morpheme.
TRUNCATION: a morph consisting in the deletion of material but still with some meaning
(laboratory – lab). It is associated to a social meaning, used in a more informal area. It has
semantic difference- LAB IS NOT THE SAME AS LLABORATORY, SOCIAlyL speaking. Also called
clipping. Keep truncation for a different process
For example,
- the past of some irregular verbs like brought (internal vowel change and suffix ‘‘t’’
4. Classification of morphemes.
FREEDOM OF OCCURRENCE
‐ Free morph: a morph that can occur in isolation. For example, cat (it’s a free morph, doesn’t
need the help of any other word, it is also a lexical morpheme), the (grammatical morpheme, it
is used to make the reference of the entity denoted by the noun, there is no lexical content).
‐ Bound morph or morpheme: a morph that can only occur in a word form in conjunction with
at least one other morph.
For example, dent‐ (root, lexical bound morph, it has meaning, it has lexical content), ‐ness
(affix, derivational morpheme), ‐s (inflectional morpheme).
FUNCTION/ EFFECT
‐ Inflectional morphemes result in different word‐forms of the same lexeme. For example, ‐s
(plural). Created different word forms of a lexeme.
‐ Derivational morphemes result in new lexemes. They belong to word‐formation domain. For
example, un‐ (negative prefix)-> uninterested.
- Class‐maintaining: morphemes have the function of indicating a particular lexicosemantic
characteristic (if derivational). For example, ‐ess (female) (lion-lioness), ‐age (collective).
b) Most prefixes are class maintaining, for example micro‐, pre‐, un‐.
‐ Class‐changing morphemes have more abstract meaning and may be seen as markers of
syntactic class, they are more abstract. For example, an affix that changes the grammatical
category into another category (noun- adj) –ful: beauty (noun) -> beautiful (adjective).
a- Adj/Adv Loud-aloud
a- V->Adj Sleep-asleep
‐ Root: that part of a word form that remains when all inflectional and derivational suffixes
have been removed. It is a lexically contentful morh-> lexical content, it has meaning (like the
root dent-). The root always have lexical content. We can’t analyse roots into smaller
elements.
For example, touch (free (can be used in isolation), it’s a morpheme), dent‐ in dental (dent is
the root, not the lexeme, bound roots do not appear in dictionaries). Bound roots are like free
roots in having lexical content, but they lack syntactic category (dent´ is nothing in itself, we
won’t be able to find it in dictionary). We don’t attach dent no anything, so it is a root.
For example, un‐ touchable Base (Complex) (-un is a unit where we attach a meaning, but it is
not a root). All roots are bases when we used them in the creation of new lexemes (a base for
a process of derivations) However, not all bases are roots, for a base to be a root need to be a
simple root. The root of this example is touch. Roots are minimal units.
Compare:
Touchable – base
The term ‘‘base’’ is used because English frequently allows affixation to forms already
containing affixes.
b) More than one root-> file cabinet, open-mended (open is a root and mind is another
root, it’s also a complex base bc it has two roots (in this case)/or having one root and one
affix)
c) A root plus one or more affixes: friend‐li‐ness- friendly is a root, it a complex base.
‐ STEM
that part of a word form that remains when all inflectional affixes have been removed or to
which an inflectional affix can be added (British tradition).
Also used to refer to bound roots or -in some theories like FDG- to free roots. To avoid
confusion, Plag proposes the alternative term ‘‘inflectional base’’.
Suffixes: attached after a base. Derivational & inflectional. Typically, class‐ changing.
c) Linking elements: which appear between the two elements of a compound (the o in
neoclassical compounds like electrolyte).
‐ Affix: a morpheme which only occurs when attached to some other morpheme or
morphemes such as a root or stem or base. Always bound. “Bound morphs which do not
realize unanalysable lexemes” (Bauer).
Replacive morphemes
We have:
‐ Final replacive morphemes: word final consonantal differences between otherwise identical
words, correlating with the differences between grammatical classes.
a) Distinctions between Noun & verb: belief‐believe (also orthographic change), mouth
(/θ/)‐mouth (/ð/), advice‐advise (final fricative), Intent‐intend (final plosive, voiceless-voiced).
b) Distinctions between adjective & verb: safe‐save.
c) Distinctions between singular & plural: wife‐wives (/f/‐/v/), path‐paths (/θ /‐/ ð/),
house‐ houses (/s/‐/z/).
c) Adj‐noun: proud‐pride.
e) Present‐Past: sing‐sang.
Sometimes, both final and medial replacive morphemes are combined: glass‐glaze, grass‐graze,
bath‐bathe, choice‐choose, breath‐breathe.
‐ Stress shift noun (in words of roman origin), (stress on 1st syllable) – verb (stress on 2nd
syllable). Typically, two‐syllable words of Romance origin. For example, ‘discount‐dis‘count,
‘import‐ im’port, ‘increase‐in’crease, ‘insult‐in’sult, ‘tormenttor’ment, ‘transfer‐trans’fer,
‘transport‐trans’port.
‐ Stress shift and vowel alternations, vowels are reduced to /ə/ or /ɪ/ in unstressed syllables.
Examples:
Zero affixation (minimum degree of modification)
The lowest possible degree of formal modification on the base. No formal difference.
‐ Inflection: zero plural (sheep), past tense of some irregular verbs (hit).
‐ Derivation: conversion (cheat (verb and noun), comment, exile, respect, preserve,
concern -> nouns and verbs).
Affixes have a more abstract; whereas -logy has a meaning, it has content. They are recurrent
elements.
Case 1: Neoclassical elements-> allow different positions, can make words on their own, very
lexical.
However, if we consider them affixes, we have to admit that words like neurology are only
made up of affixes, which goes against the definition of affix (attached to a root or base). We
cant say that a word if formed only by affixes.
In addition, other cases can be found in both initial and final position, unlike affixes (derm‐
atitis, endo‐derm). We don’t have affixes that can be attach at the beginning.
They have higher degree of lexical content (root?). Compare neuro‐ “pertaining to the nervous
system or brain”, re‐ “again”. Though the distinction is mainly based on the degree of lexical
content (or lexical density), this is not always easy to determine mini‐, super‐, mega‐.
These prefixes convey notions of size that are often conveyed by affixes in other languages.
Conclusion: different morphologists might draw the dividing line between bound roots and
affixes in a different place.
Case 2: Splinters
Analyse the following words into their component elements: alcoholic, workoholic, chocoholic,
Potterholic, alcohol‐ic, work‐oholic, choc‐oholic, Potter‐holic, watergate, Monicagate,
nipplegate. -Oholic initially was a part of a work.
A splinter is a portion of a word that is non‐morphemic to begin with (?) but has been split off
and used recurrently on new bases, that is, a “brand‐new” affix. Semantically, they carry some
of the semantic content of the original word from which they split (“addicted to”, “political
scandal”), and therefore are more contentful than typical affixes. Formally, they can be:
‐ An affix plus some portion of the original base: choco‐holic (from chocolate
‐ The end of a mono‐morphemic word: cashorama (from panorama, I guess, not found
in web).
Case 3: Rankshift
Affixes used as if they were free. They have undergone rankshift (Halliday) She drove a
mini / That’s super / Pros and cons / ‘Was it good?’ ‘Ish’.
Test, caps
Ashes
Uncontrollability:
Control- root
Uncontrolabil-base, -ity-suffix
Alternative analysis:
Control- root
“Lack of controllability”
2. Postcolonialism
Colony-root
Alternative analysys:
Colony (root)
Post(prefix), colonial(base)