See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/233050828

A Holon Approach to Agroecology

Article  in  International Journal of Agricultural Sustainability · January 2007

DOI: 10.1080/14735903.2007.9684828

CITATIONS READS

50 249

2 authors:

William L Bland Michael M. Bell

University of Wisconsin–Madison University of Wisconsin–Madison
56 PUBLICATIONS   1,541 CITATIONS    68 PUBLICATIONS   2,011 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Masters thesis View project

The Agroecological Imagination special commentary section View project

All content following this page was uploaded by Michael M. Bell on 22 May 2014.

The user has requested enhancement of the downloaded file.

# 2007 Earthscan www.earthscanjournals.com

A holon approach to agroecology1

William L. Bland1 and Michael M. Bell2
1
Department of Soil Science, University of Wisconsin-Madison, Madison, WI, USA; and 2Department of Rural
Sociology, University of Wisconsin-Madison, Madison, WI, USA

Systems thinking contributes to envisioning agricultural sustainability. However, it faces two dilemmas,
recently highlighted by complexity theory: the problems of boundary and change. We propose that
interpreting Koestler’s holon as an intentional entity embedded in an ecology of contexts provides an
ontological construct which addresses both of these issues. The holon is in some ways a whole and
in other ways a part, and to see it simultaneously as both we suggest an epistemological tool that we
term flickering. In our interpretation a holon is bounded by its intentionality to persist, and the
imperative to do so in multiple, incommensurable, and ever-evolving contexts motivates – indeed,
makes both possible and inevitable – change. Farms are compelling examples of holons, as their
humans plan and act to maintain them as a source of livelihood, necessarily in contexts as diverse
and shifting as climate, life histories, trade rules, subsidies, personal spirituality and public
perceptions of agricultural practices.

Keywords: systems, complexity, theory, agriculture, sociology, human-environment systems,

agroecology, holon

Like all major human endeavours (our lists would along the way, learning to welcome the perspec-
surely overlap), agriculture has pervasive and tives and contributions of a great many entry
deep connections with diverse issues. In the case points for the study of agriculture, from soil
of agriculture, the prosperity of the vast majority science to social science, from agronomy to
of species, myriad human cultures and our own zoology.
spiritual sensitivities are all shaped by, and in The turn to systems thinking has been decisive.
turn shape, these connections, these involvements. Agricultural theorists embraced systems thinking as
Agriculture is a huge, and hugely important, under- a powerful and essential epistemological tool. Early
taking, and its students are at long last finally articulations of the interconnectedness of the bio-
coming to appreciate this involved hugeness. This physical and social, for example Bawden and Ison
is a great step forward from the many decades of (1992), Conway (1987), Pearson and Ison (1987),
understanding it more as an ensemble of discrete looked to systems thinking to structure further
bits – a large ensemble of bits, to be sure, and research on farming systems in developed settings
one that is articulated here and there, but still as toward efficiency, and in more resource-poor set-
a kind of machine, fundamentally separable part tings to make more effective interventions leading
from part, and as well from the world at large: to greater productivity for vulnerable farmers and
rigid, linear, detachable and controllable. We are, populations. Thirty years later, systems thinking
serves as a fundamental tool in the task of transform-

Corresponding author. Email: [email protected] ing agriculture towards a ‘sustainable’ future

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 281

(e.g. Gliessman, 2004). It has now become a matter observations are part of very different stories, for
of our intellectual reflexes to speak of agricultural example Bill Cronon’s (1992) comparison of
systems, to consider the ecological context of agricul- alternative tellings of the meaning of white
ture, to invite participation from stakeholders into migration into the US Great Plains reveals it to be
agricultural decision-making, and to embrace the interpretable as destruction of peoples and nature,
complexity of doing all of these. We are seeking or the triumph of a people over nature. The ‘soft-
interconnections, we are finding them, and we are systems’ methodology of Checkland (e.g. Check-
doing better at acting on them. land & Scholes, 1999) confronts directly the narra-
All this is good news, but there remain a number tive challenge in the boundary problem. They
of places for intellectual growth in this more argued that systems thinking always in part reflects
involved understanding of agriculture, particularly the priorities of the systems thinker, and has to be
with regard to systems thinking. As valuable as it understood as caught up in human institutions
has been for conceptualizing agriculture’s and politics, with all their implications for ideology
complex involvements, there are limitations in and social power. Checkland (Checkland & Scholes,
what systems thinking allows us to describe and 1999; 7) identified diverse narratives connected to
discuss. Our principal concern is that the language the program to build the Concorde supersonic airli-
of systems encourages an over-connected under- ner beyond fabricating a machine: an important col-
standing of the world – an understanding that laboration between the British government and the
leads to two dilemmas currently besetting systems country’s leading aircraft manufacturer, a project
thinking: the problems of boundary and of change. to stimulate European engineering, and an exercise
In the under-connected view of agriculture – as a in British collaboration with European partners.
collection of tools and techniques to be steadily Thus the narrative problem is about choosing
improved – that formerly held the intellectual which story is being told about a set of events,
field, boundaries were precise and secure, if because there are always multiple possible mean-
unrealistic. But if as the systems view would have ings to any situation. Selecting a story forces the
us see it – that all is connected – how do we analyst to seek an ontology, that is, to make
draw boundaries by which we might understand decisions about what entities and relationships are
the world, as William James phrased it, as anything important and must be emphasized, in order to
other than a blooming, buzzing confusion? Where tell this story. While the subjectivity issue is an epis-
are the surfaces and breakpoints of significance? temological question, the narrative issue is onto-
Where does, for example, a farm begin and end? logical. Unfortunately, this means that systems
At the property boundary? At the edge of the thinking must represent only a partial view, in
watershed and the wildlife migration corridor? At two senses of the word: limited and, in the end, pol-
the consumer’s dinner plate? Indeed, if everything itical. The question at hand is limited by the (epis-
is connected, can there be a surface or breakpoint temologically essential) boundaries that, alas,
that bounds any thing? All is one, and the analyst allow us to address only part of the larger web of
is immobilized in finding an intelligent way to connections, and it is made political, that is, a
describe, research, and account for particular por- matter for contention, by the ontological choice
tions of the system. Yet in order to proceed the of which questions to ask and which to not ask.
domain of the analysis must somehow be limited An early exposition of systems thinking in agri-
or bounded, explicitly or otherwise. This is an epis- culture provides an opportunity to reiterate the
temological challenge that we term the subjectivity subjective and narrative boundary problems. In
problem of system boundary. There is an inevitable Spedding’s (1988) An Introduction to Agricultural
subjectivity in defining the system to be studied, Systems, he illustrates the subjective boundary
that is, another analyst might well draw the bound- problem by considering the mass and energy
aries demarcating the system of interest differently. exchanges of a chicken, and specifically the impli-
An equally vexing challenge is deciding on which cations of confining said chicken in a box. Once
of the many stories that might be told about the so confined, the boundary of any systems analysis
system are we trying to tell – what we term the nar- about the chicken’s physiology probably must
rative problem. This arises because the same set of expand to include the box and perhaps the box’s

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

282 W.L. BLAND AND M.M. BELL

environment. But maybe not, depending on the means of conceiving any incompleteness of involve-
particular question and assumptions about what ment, any options for the emergence of new entities
aspects of the chicken will change as a result of and connections. Yet that we should be prepared
being in the box. This is the subjective boundary for infinite, unimaginable possibilities is surely a
problem: where can we validly draw the outer fundamental lesson of Darwinian evolution. Ways
boundary of our system? Few situations are any- to talk about these infinite possibilities are essential
where near as simple as this example. Spedding to the possibilism and unpredictability of a deeper
did not address the narrative boundary problem, sense of the meaning of change.
however. Here we imagine the other stories in Further a system model is inevitably obsolete.
which the chicken plays a role. Does placing it This out-of-dateness arises because it can only be
in a box raise ethical issues about using animals constructed from observations made over some
in research? Was the farmer who raised the span of time. Our presumption of change means
chicken rewarded appropriately? Did related that at least theoretically the relevant entities
farming activities degrade soil and water resources? are in flux – new types appearing, new relation-
Will the chicken become part of a nutritious meal, ships replacing obsolete ones, all the while as the
or actually harm the health of the eater? observer gains and loses sensitivities. In the snap-
The boundary problems of systems thinking shot portrayal there is an inescapable assumption
have long been recognized (e.g. Churchman, of a ‘system’ that it is just that: an identifiable
1979). Theorists continue to the present to thing, and not some other identifiable thing, static
advance procedures for addressing the problems and worked out, staying in equilibrium, despite
in an orderly and transparent way. Midgly dynamism.
(2000), for example, cites four earlier examples of The under-connected view of agriculture had a
efforts to overcome inevitable and confounding comfortable theory of levers-and-knobs change
boundaries in systems analysis, and proposes yet that gave us a pleasant feeling of control over a sim-
another. There is likely no escaping the dangers plified world. We rightly reject that linear and
of the inevitable boundaries in systems thinking, mechanical view today, but systems thinking does
only ways of being aware of their possible impli- not reject the pleasant feeling of control over
cations in applications of this epistemology – as change. On the contrary, systems thinking typically
Ulrich (1993) points out, ‘The “right” boundary presents itself as a better source of control, one that
judgments depend on the subjective interests, takes into account the real connectedness of life.
values, and knowledge of those who judge. . . But in continuing this mode of control, it has still
[and] will tend to be disputed.’ found little desire for accommodating the plain
Change is also difficult to discuss within systems reality of unpredictability (Bell, forthcoming). In
thinking and its view of a connected-up world. this way, the problem of narrative also extends to
Indeed, most systems accounts do not discuss the problem of change. In order to tell a story of
change, but rather present a boxes-and-arrows control, systems thinking has distanced itself from
snapshot, with perhaps a tip of the hat to highlighting the incompleteness, disjunctures and
‘dynamic equilibrium’. But even dynamic possibilism that undermine such a story.
equilibrium is a kind of change without change in A sense of control is supported by a feeling of
that, under the presumption of connectedness, stability in the boundaries we draw. A rich sense
everything is accounted for and understood. It is of change brings us back to the narrative challenge
predicted change, and thus in a deeper sense is of boundary. The most powerful and least unpre-
not change at all. As well, there is a presumption dictable changes are ontological: when the import-
that the connectedness of things puts them at ‘equi- ant entities and their connections seem to have
librium’, a kind of constantly readjusting balancing somehow shifted. The result, as we will describe,
act in the involved hugeness of it all. But is the is a tendency of systems thinking to present an
world everywhere balanced out, or even seeking overly tidy view of agriculture, in which the very
balance? Are there no disjunctures, conflicts and collection of parts that are relevant, and the bound-
contradictions? Does it really function as some aries we draw through them, are too precise, too
great whole? Systems thinking gives us little stable, and too worked out. Further, we may see

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 283

the connections within as stronger and more rigid planning humans seek configurations of their
than they really are, causing us to fail to imagine endeavours that will allow them to sustain their
myriad alternative behaviors. Doubts have been agricultural ambitions. It is, we believe, important
growing about the advisability and feasibility of to start out with a reverence for this planning.
system theory’s pleasantness (e.g. Allen et al.,
2001; Checkland & Scholes, 1999; Rosen, 1991,
2000; Vayda, 1986). The term ‘complexity Systems thinking and the problem of
theory’ is perhaps the best overall summary of over-connectedness
what these authors have tried to highlight for us,
in their various ways. ‘When we try to pick out anything by itself, we find
We describe here a path that leads on and out it hitched to everything else in the Universe’, wrote
from the insights of complexity, and especially John Muir (1911), a much-cited line, and justly so.
from a reframing of Koestler’s (1967) idea of the Its imagery aptly captures the anti-reductionist
holon. We adopt the holon as a key ontological stance that has long been a part of the ecological
type, and an analytic method that we term flicker- mind, and that has always characterized systems
ing as an essential epistemological tool. We will thinking. This is often a good emphasis to have.
describe how the holon enables discussion of Systems thinking provides a language and habits
boundary and change within a world of involve- of the mind that repeatedly alert us to the idea
ment à la systems, while flickering gives us a flex- that ‘you can’t do just one thing’ (variously
ible vision of boundaries that remain open to ascribed to Leopold, Bateson, Ehrlich, Campbell
evolution and unconnectedness. and no doubt others). There are typically numerous
We apply the notions of holon and flickering to implications of an action, some of them distant
agroecology, a word we greatly admire because in time and space, and the under-connected vision
its etymology asks us to consider the issues of agri- of reductionism encouraged us to overlook
culture’s connections. A scholar-student may these, at least at first, often to our eventual
approach agriculture with little more than a sense dismay. As we have come to rue the stubbing of
that it is a rich and fundamentally important our toes, we have come to embrace the
complex of activity, worthy of study, or with a par- word system.
ticular problem in mind, say, protein malnutrition These are old intellectual troubles, however.
in Sub-Saharan Africa, or the relatively high cost Although it has come to seem a relatively new
of growing apples in Wisconsin. With either such concept, with the continuing flurry of academic
a general or specific concern in mind, the agro- writing on it since the 1950s, the word system actu-
ecological commitment, in our understanding of ally dates back to the ancient Greeks. It was they
it, is to seek to transcend any one particular entry who combined the root syn, meaning ‘connect’ or
point into the agricultural debate. But to seek ‘combine’, with histanai, meaning to ‘set up’ or
such a transcendence should not be to presume ‘establish’, into sustema, in order to describe an
entry points do not or should not exist. If all the ‘organized whole’ (Onions, 1955 (1933)). The
world is connected, then there are no connections Romans felt they needed to be reminded of these
to make, nothing to transcend, nothing to learn. insights too and they took sustema almost directly
Thus, we offer the holon approach not as a final into Latin, calling it systema.
answer, but as a way for the agroecologically But there are perils as well in carrying this sense
concerned to at least agree on a radio frequency of an organized whole too far. These too are perils
on which they might communicate with one that have long troubled us. Aristotle, for example,
another as they pursue their journeys through railed at the ‘monism’ he found in the thought of
complex agricultural questions. his teacher, Plato, who offered a united view of
The holon approach, as will become clear, takes the world in which ‘the Good’ created all things
as one of its points of departure that agriculture is, and all things were a manifestation of ‘the Good’.
most fundamentally, humans planning and acting With such a perspective, Aristotle (1987) retorted
to cultivate a livelihood through the phenomena in the Physics (185b: 15– 25), all things ‘will be
of plant and animal increase. Further, these the same, and the thesis under discussion will no

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

284 W.L. BLAND AND M.M. BELL

longer be that all things are one, but that they are consequence, such as recycling. To do agriculture
nothing at all’. is to do many things, not one alone, the figure
Systems thinking, in its currently common rightly tells us, as it is indeed hitched to much
modes, courts these Platonic troubles. Let us take, else. Moreover, the figure plainly speaks to the
for example, a recent effort to present an overview need for more than one disciplinary voice to
of the agricultural endeavour from a systems point handle this involved hugeness. This is all to the
of view, by the noted agroecologist, Steven Gliess- good.
man. We reproduce in Figure 1 the visual represen- But let us next note some matters that this figure,
tation that Gliessman gives of his systems analysis, and approach, does not easily alert us to. First,
what he terms the ‘functional and structural com- there is its neat and tidy appearance, with precise
ponents of an ecosystem converted to a sustainable lines and boxes and arrows and feedbacks, drawn
agroecosystem’ (Gliessman, 2004: 21). Let us with computerized exactness. Second, there is the
immediately emphasize some of the helpful features language Gleissman’s article uses to describe this
of Gliessman’s figure, most notably its emphasis on view of agroecology – that it is based on ‘the
agriculture as having a broad array of connections balance needed for long-term sustainability’, that
with human endeavour, including aspects generally it is ‘a functional system of complementary
not thought of immediately as having agricultural relations’, that it strives for ‘equilibrium’, albeit a

Figure 1 From Gliessman (2004): the ‘functional and structural components of an ecosystem converted to a sustainable
agroecosystem’

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 285

‘dynamic’ one (Gliessman, 2004: 19– 21). Similar comprehensible and programmable, and easier to
language shapes the depiction of an agricultural act on. Weather forecasts, for all their known and
system in Spedding (1988) as ‘operating together likely irresolvable inaccuracies, help billions
for a common purpose’, or as ‘integrated to accom- through their day. But the tidiness of a simple
plish a well-defined purpose’ (Peart & Shoup, systems model will in most circumstances involve
2004: 2). Like Gleissman’s diagram, words and a fair bit of sweeping under the rug.
phrases like complementarity, equilibrium, balance, In simple systems, then, we assume we have
common purpose and integration portray agro- knowledge of all of the relevant parts and their
ecology as orderly, articulated and unified: as clean. interconnections and interactions, while in com-
But is the world a neat and tidy place, at least always plexity we take it as a starting point that this is
and everywhere? In a connected world, how can not possible. In philosophical terms, then, epistem-
we draw such precise boundaries through reality? ology and ontology are equal in simple systems,
And where is there scope for change, beyond a that is, we know about the system (epistemology)
dynamic return to equilibrium? Is agriculture so as a result of the fact that we chose what entities
finished, complete, and balanced out? and relationships to include (ontology) in our
Complexity theory would argue that it is not, simple representation (Rosen, 2000: 281). The
and that systems thinking is as much an effort to numerical simulation models widely used in agron-
create a sense of order as it is to find it. The term omy are simple in this sense. They are viewed by
‘complexity’ is used both colloquially and formally their creators and users as generic descriptions of
in several ways. As well, some theorists have made phenomena of interest, for example, how seeds,
broadly equivalent arguments without invoking the soil, water and sunlight interact to bring forth a
word ‘complexity’ as their central term, such as crop. Comparisons to actual observations are
Checkland and Scholes’s ‘soft systems method- imperfect presumably because of some special
ology’. We use ‘complexity’ in part as a shorthand characteristic of the harvested test plot. This is
to differentiate this broad body of work from the exactly the opposite of Rosen’s perspective, that is,
tidy view of systems, in much the way that Rosen that reality is complex and the generic situation;
(2000) does. Rosen begins with the view that all for Rosen, the simple model, because we have
reality is complex, in the sense that any given made choices about what to exclude, is special and
action has the potential to cause unexpected therefore of limited applicability (Rosen, 2000: 304).
results in unexpected places. For Rosen (2000: Allen et al. (2001) have proposed a more fully
306), ‘a system is called complex if it has a nonsi- postmodern approach to the selectivity of system
mulable model’. In other words, a complex model thinking, arguing that systems are always stories
expects unpredictability. In this sense, the more inevitably told from the perspective of a storyteller,
accurate model is a less accurate model. and that, properly understood, this is not necess-
One could, of course, deliberately choose to arily cause for intellectual alarm. They point out
study a portion of reality by imagining it as what that narrative allows the storyteller to adroitly
Rosen (2000) has suggested terming a ‘simple leap over manifold scales of space and time when
system’ – that is, as a collection of entities and describing the web of involvements an action
interactions whose behaviors we can ‘model’ might entail. This expansion/compression of time
through equations and algorithms and thus and space greatly reduces the care required in defin-
predict with some acceptable amount of error. ing system boundaries, if not negating it altogether.
Such simple system models may be highly detailed, Here the analyst need no longer support a conjec-
for example those used in weather forecasting – ture with, for example, output of a meter-scale
numerical models of the atmosphere are among hourly timestep model purportedly simulating con-
the most sophisticated and computationally inten- tinents over centuries. This narrative approach
sive simulations that exist of any physical (correctly) understands that this is not likely valid
domain. They are nonetheless simple in the sense anyway, and rather draws on whatever arguments
that they presume an ordered predictability. are at hand for looking ahead. But as we look
There may be considerable utility in having such through the recent pages of our agroecological
a model, for the result is more easily journals, we see little explicit use of such a

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

286 W.L. BLAND AND M.M. BELL

postmodern take on complexity and narrative. it is essential to envision each simultaneously as a

Most authors continue to use a simple systems whole and as a part of other entities. An individual
ontology and epistemology. For them perhaps the carries on physiological functions as a whole, yet
postmodern view is intellectually alarming, or exchanges food and waste as part of an ecology.
maybe it seems more a matter of philosophical The academic department carries out many func-
truth than the practical truth they seek. tions autonomously from the university in which
We thus detect a nervousness in the current it is embedded, often to its regret and frustration,
moment: while it seems plain now that the confi- yet has little meaning outside the context of the
dences of reductionism were misplaced, and that a larger institution, which has its own autonomous
more connected view is necessary, a deep dilemma movements. Switching to an agricultural frame,
has opened up in agroecological thinking. Simple certainly an individual animal reared as part of an
systems are solvable but wrong, while complexity agricultural endeavour may at times be usefully
theory is unsolvable but right. Neither seems a com- thought of as a whole, that is, it most probably
fortable position to maintain, but perhaps through has intent and capability (about which we will
appeal to both schools of thought we can develop have more to say later), and perhaps moral
new conceptual tools to help us think about and standing in its own right, but is also part of a
act creatively within human-environmental relation- farm. We may say as much (if not more) of the
ships like agriculture. farmer himself or herself. Both animal and farmer
are part of the farm holon, while themselves
being wholes comprised of parts, just as the farm
From systems thinking to holonic is part of larger holons, such as the agricultural
thinking economy.
A holon exists within an ecology of contexts.
We would like to suggest a theoretical framework These contexts collectively form the situation in
by which agroecologists might better accommodate which the holon functions. Contexts important to
the subjectivity and narrative boundary problems, the farmer might include, for example, family,
as well as the crucial issues of innovation and farm business, genetic heart disease, and spiritual
change. We propose drawing into agroecology a beliefs (see Bawden & Ison, 1992 for a rich
number of ideas from complex system thinking, compilation of the breadth of the issues that
specifically a reconceptualization of holonic think- imaginably enter here). The holon is a nexus of
ing. We first introduce the holon and the concep- many contexts, involved in infinite ways with
tual aid of flickering. them, yet still an identifiable entity. Indeed we
only recognize anything as a separate entity if it is
somehow visible against the background of its
Seeing holons requires flickering
contexts – it is the contrast with these contexts
The term holon was proposed by Arthur Koestler that give the object of our attention any identity
(1967) to address the problem that interesting enti- at all. A particular context of a holon may itself
ties in biology and society are in many senses be a holon, but not necessarily (a point we will
wholes, but, on the other hand, can not be under- return to).
stood without recognition of contexts in which These contexts are incommensurable, that is, they
they survive. For Koestler holons are entities that cannot be compared directly to one another, or
have autonomy in some senses, yet are clearly a converted to a common unit of measurement,
part of something larger in other senses. As Koes- despite the best efforts of economists. They are as
tler (1967: 210) put it, ‘Parts and wholes in an diverse as the farmer’s beliefs about the sentience
absolute sense do not exist in the domain of life. of livestock and the cost of corn. An important
The concept of the holon is intended to reconcile implication of incommensurability is the impossi-
atomistic and holistic approaches.’ Clear candi- bility of calculating optimal configurations of the
dates for the holon label include individual holon. Determining such optimal arrangements
humans and university departments: to answer requires the ability to mathematically trade a
many questions that one might raise about them bit more of this for a bit less of that, but

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 287

incommensurability makes this impossible. At the whole). Thus parts require wholes, wholes require
scale of the individual farm it may be a choice parts, and yet neither pure parts nor wholes actu-
between growing more corn or creating habitat ally exist.
attractive to grassland birds, while at the inter- There is, however, no necessary discomfort in
national scale it may be low-cost shrimp at the this paradox. The trick is to learn to continually
cost of mangroves. switch back and forth between the perspective of
A holon thus bounds a collection of entities that the part and the perspective of the whole, some-
are involved with one another relatively intimately. thing we call flickering. The imperative of flickering
This close involvement within a holon suggests that arises because, for most of us, our minds seek to
a change, movement, or action somewhere within settle on a single representation of an entity.
the holon does not happen in isolation of the rest Think of the well-known outline drawings that
of the holon, for example, a major loss of health appear to be of two different depictions, depending
by one member of a farm family likely has fairly on very small changes of focus. In one of these,
immediate repercussions throughout the enterprise. many observers first see a goblet, while others see
In contrast the holon is less intimately involved two faces in silhouette. Our vision tends to settle
with the broader ecosystem of contexts in which on one or the other and must be consciously
it exists, and not all changes in a context will pulled to see the other. So the patterns of white
necessarily impact the holon appreciably. While and black are (at least) two distinct images, just
the farm holon is likely impacted by the health of as a holon is both a part of something greater and
each member of the household, it is far less clear a whole in its own right. Flickering gives us a tool
that the health of the top government official in a with which we can engage the paradox that
country’s department of agriculture would have holons are simultaneously wholes and parts (the
significant implications. But it might – any involve- subjectivity boundary problem) and players in a
ment leaves open the possibility that the effective- set of sometimes competing stories (the narrative
ness of this distant official might be significant to boundary problem). Further, while holons are
the farm, if government policies changed as a both part and whole, they are not completed, com-
result of his or her illness. pletely worked-out manifestations of either. The
Allen and colleagues (Ahl & Allen, 1996; Allen whole is always reshaping its parts, the parts that
& Starr, 1982) further elaborated the holon constitute the whole are ever changing, and thus
concept, emphasizing, as did Koestler, a presumed so is the resultant whole, and so the very bound-
hierarchical organization of living and social aries around what might be identified as a part
systems. For us hierarchical organization is are transient. Flickering helps us imagine the cogni-
helpful for building the notion of holons (as evi- tive ‘light touch’ that allows us to remain open to a
denced by our earlier examples), but the holon fuller range of possible interpretations of things and
and its ecology of contexts is much too messy to events. Flickering gives us a glimpse of the holon’s
be usefully envisioned solely in hierarchical terms. transcendence of part and whole – an ephemeral
Rather, our development of the concept of holon state between these two far more concrete ideas –
takes seriously Koestler’s suggestion that both as the flickering light is between on and off.
parts and wholes do not exist in an absolute The point here is not to argue against ever
sense. Systems thinking, even when embracing drawing a boundary – to do so is to put aside
complexity, has continually placed its emphasis thought. But we need to see an agroecological
on the notion that parts do not absolutely exist – boundary as a kind of two-ness, not a hard singu-
that they are always connected to something else, larity. Envisioning a holon’s bounds too tangibly
and that one can never do, or be, one thing. This risks atomizing the situation, while too ephemeral
boundless view, while a vital insight, all too easily a boundary leads to complete dissipation of the
slides into the view that all is an appropriately con- topic and thus toward a totalizing holism. Flicker-
nected whole, that is, functionalism. Our case is for ing keeps this problem ever in our minds. We need
recovering Koestler’s implication that we should this conceptual stereo view, with one eye for part-
equally interrogate the manifestations of wholeness ness and one for wholeness, to avoid conceptual
in what is readily seen as a part (of some greater stereotypes. We need a continual shifting back

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

288 W.L. BLAND AND M.M. BELL

and forth in our depth of focus: every whole a part; (1996) suggestion that a ‘robust’ boundary is one
every part a whole; every whole and part neither. It that remains useful from multiple prespectives, or
is not intellectually uncomfortable to do so. Indeed, ‘observation regimes’. Such multiple perspectives
we submit, this flickering is the most every day of arise with both the subjective and narrative bound-
acts, as we negotiate our own lives as parts, ary problems. Perspectives might be variations on
wholes, and neither. what physiological phenomena can (or must for
the work to be valid) be included in research on
measuring animal welfare – a subjective boundary
Intentionality bounds holons
issue – or on what story about the role of animals
We have proposed that the holon is a useful con- in agriculture we are trying to tell – a narrative
ceptualization of a to-a-degree bounded entity boundary issue. By the criterion of utility from mul-
within a web of involvements. Its usefulness for tiple perspectives farms are appropriately holons,
addressing the boundary problems, however, for they are, at least, sources of livelihoods for
turns critically on our ability to decide what is owners and workers, centres of economic activity,
and is not a holon and what is included in some- producers of food, and large-scale manipulations
thing so identified. We propose that intentionality of land, air and water resources.
is the primary criterion for identifying and bound- A holon may have capability to affect change in
ing a holon. By intentionality we mean the active some of its contexts, but not in others. Here we
envisioning and seeking of a set of goals, such as have in mind a holonic take on Amartya Sen and
the farm family working and planning so that Martha Nussbaum’s vision of a person’s ability to
they may continue to derive a livelihood by collect- be and to do (Nussbaum & Sen, 1993; Sen, 1992,
ing milk from cows. Active intentionalities in the 1999). In the language of agency, we mean that
world seek to maintain themselves as wholes of an intentionality’s agency towards something
mutually involved parts, and this usually requires must also be understood in the context of its
that they also try to maintain themselves as parts agency from, its degree of release (Bell, forthcom-
involved in wholes, through their flickered imagin- ing). Regardless of how hard a farmer tries to act
ation of themselves and their contexts. The humans intentionally towards the rain, he or she lacks capa-
in a farming enterprise try to maintain the farm as a bility over the atmosphere. Conversely, the choices
whole amid the colliding disjunctures of each a farmer makes about tillage, manure and crop
passing day, and do so in part by trying to maintain rotations can affect soil, expanding or contracting
the farm as a part of markets, cultures, and ecol- the possibilities this context provides. Understand-
ogies that may not integrate with the farm as ing the capability (or lack thereof) a farmer has
closely as those humans might like. They try to over specific contexts is an essential task for both
get along better with each other within the farm, the farmer and the agroecosystem analyst.
just as they try to have the farm get along better Farmers can ill afford to tilt at windmills, even if
with changing prices, values, and rainfall. The the agrotechnologist has scientifically determined
potential for a better state to get to comes from a that it is a new and better way to do agriculture.
holonic recognition that no whole or part is just On the other hand, a rich appreciation of the possi-
that, a part or whole over and done with, and bilities for beneficial positive feedbacks from
that wholeness and partness are always changing. improved soil health can pay dividends.
Intentionalities strive for and act on these poten- Appealing to intentionality as a criterion for
tials, and in the process create them. At the same identifying holons leads us to see that some com-
time, contexts shape the direction – the sense of ponents of the farm (or any other) holon are them-
motivational pull – of intentionality toward a selves holons, while others are not. Similarly, some
vision or plan (Emirbayer & Mische, 1998; of the contexts in which a holon exists may them-
Martin, 2003). Intentionality emerges out of the selves be holons, but not all are. What a farmer
contexts in which it seeks to act, possibly trans- can do and be is surely shaped by the tractor, and
forming them. thus the tractor acts upon the farm and the
Looking toward intentionality as a guide to the farmer, and is not merely pushed around by the
boundary problem appeals to Ahl and Allen’s farm and the farmer. The tractor has capability, it

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 289

has consequence, and it acts in the world just as ecology of contexts. The totality of the holon and
humans do, as the sociologist Latour has so its contexts is an involved hugeness, and yet unfina-
copiously argued. But any intentionality for the lized. If we wish to conduct research in order to
tractor must come from beyond its own surface. effect change we must draw some boundaries
Indeed, it is that external intentionality that within this hugeness, and intentionality offers a
makes it a tractor to begin with, and not, say, an surface that is imaginable, appropriately poorly
interesting work of iron sculpture to be admired defined and porous in spots, and of great signifi-
in the front lawn, or a place to build a nest for cance to the persistence of the sorts of entities we
the new litter of ratlings. care about understanding more richly.
Not all boundaries of consequence are holonic
boundaries. In our development of the concept of
Representing holons
a holon, a holonic boundary is a boundary that
intentionally tries to maintain itself as such, as a Many of us would find useful a visual rhetoric for
surface of consequence, across the changing describing holon agroecology. There is danger
dynamics of its situation. The tractor does not here, as the stability of the inscribed page implies
maintain itself (much as a farmer might wish it the very completeness and settlement that we
would), not as a tractor nor a sculpture nor a nest wish to keep forever in question. We elect to
site. It is the holonic intentionalities of the world attempt the visual, deciding in favour of perhaps
that give the tractor’s particular capability its serving a broader community, at the risk of
specific consequence. Or not. A tractor, then, is making too concrete our concept of the holon.
not a holon. Describing our visual representation will also give
By contrast, take a mule. Intentionalities external us opportunities to address and develop several
to the mule might find very different consequence in implications of a holonic approach, including con-
it. A farmer might see a form of animal traction, a texts, incommensurability, change, and flickering.
child might see a dangerous threat, and a fly might Figure 2 shows a farm holon from two perspec-
see a source of nourishment. But apart from what- tives. In the centre of each view we show a semi-
ever these external intentionalities might find of con- distinct entity – the farm holon – with an irregular
sequence in the mule, the mule will act on its own surface and with many internal entities, also with
with regard to its bounding surface, seeking to similarly irregular internal surfaces. These entities
keep that surface as a persisting, yet fluxing, do not necessarily quite fit together, and we have
source of consequence. A mule, then, is a holon. tried to show them with overlaps and disjunctures
As we have noted, some of the contexts in which (which are more apparent in a color version of
our holon exists are themselves holons, manifesting the figure). But they are acting, perhaps sloppily
intentionalities. As holons go about the work of per- and disjointedly, to create and maintain the farm
sisting they ignore the intentionalities of other holon. Some of these internal entities are them-
holons at their peril, for example, the farmer selves holons, for example, farm family members,
should realize that the banker seeks to minimize hired labourers, and farm animals, while others
risk and maximize profit by charging higher interest do not constitute a surface of intentionality on
to those who can least afford it, or that transnational their own, and thus are not holons.
food companies seek the lowest-cost supplies of raw In the upper panel the holon creates a nexus of
material. Similarly, external analysts, for example, the contexts in which, and with which, it must con-
agroecologists, will miss a good bit of the story if struct and continually reconstruct itself. We show a
they fail to recognize and acknowledge intentional- few – family and finance, the crop environment,
ity in the holons they study. Finally, a great variety markets and subsidies and spiritual beliefs – for
of intentionalities in the world are themselves illustrative purposes only. There is no limit to
acting from concepts of the farm as a fundamental these, either empirically or conceptually. Some con-
construct, helping justify our frequent contention texts may act holonically on their own, and from
that the farm is a useful and compelling holon. that point of view it would seem that a wise
Summarizing our argument to this point, then, analyst, and the wise holon, would recognize that
the holon is an intentional entity embedded in an characteristic. The state, for example, is a context

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

290 W.L. BLAND AND M.M. BELL

holon also extend outside of it. The little irregular

shapes we see in the holon are, in most circum-
stances, cross-sections of entities that extend
beyond the farm, and often well beyond it. A
child may labour on the farm at the weekend, and
never leave its confines, but on Monday takes the
bus for school, and thereby contributes to the con-
struction of a school as holon. A farm animal may
live all its life within the legal property boundaries
of the farm, but its genetic history and its conse-
quence for markets and the cleanliness of the
local water supply extend far beyond into other
holons and non-holonic contexts.
Each context depicted in the upper panel of
Figure 2 is a bundle of interactions between the
holon whole and the environments in which it
exists. The lower panel unpacks what we have
chosen to bundle as the ‘crop environment’
context of a farm: past cropping, hydraulic charac-
teristics, and so on. But any of the parts of this
bundle is a context that could be elevated such
that it appeared in the upper panel. The analyst
has freedom to chose the contexts relevant to the
question at hand, and is ever at risk of overlooking
Figure 2 The farm is depicted as the three-dimensional,
roughly spherical body at the centre of the upper and lower a context that is dramatically shaping the farm
panel images. The upper panel depicts the farm holon holon. The arrows in the lower panel remind us
existing simultaneously within four incommensurable sets that interactions between the farm holon and its
of contexts. The lower panel depicts the farm holon contexts are multidirectional. The farm has various
embedded in but one of these, its crop environment. This
degrees of capability with regard to various con-
set of contexts is represented as a two-dimensional space
(a plane in the page), in which the farm we wish to study texts. Toward some the farm has essentially none,
must operate. That is, whatever it is that the farm does it for example, climate, while toward others it may
must do successfully within the facets of the crop have considerable capability, for example, debt.
environment context (e.g. rainfall climate, soil hydraulic Contexts might always be said (perhaps trivially)
characteristics) of its geographic location. These contexts
to have capability over the farm, or we would not
are illustrative only – for particular analyses a different array
would likely be appropriate. have recognized them as relevant in the first place.
The holon and its contexts in the graphic depic-
tion of Figure 2 are collectively the larger whole
for probably every farm holon in the modern world that systems thinking attempts to map. The farm
and one that, holonically, tries to maintain its is a part of this larger whole, but is itself, in mul-
boundary. The climate, however, may be a simi- tiple ways, a whole – a holon. Here the importance
larly pervasive context that, while perhaps at of flickering arises, as it helps us to see the whole/
times helpfully imagined as an intentional entity, part nature of the holon. The holon is in many
as in myth and story, does not try to maintain important ways a whole, but it is also shaped by
itself as a holon. and helps construct its contexts. Our flickered
In any event, it is crucial to note in Figure 2 that imagining of this dual nature helps us envision
the contexts extend into the holon. Markets are not the many-dimensional and porous boundary of
separate from the farm holon. That is why markets the holon. In turn, this virtual boundary gives the
matter for the farm holon, and why farm holons holon (and our conceptualization of it) the
matter for the market. Many (if not all) of the enti- freedom to be the animated, ever-in-flux entity it
ties that we represent as being internal to the farm must be in order to persist. Here we arrive at the

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 291

problem of change, and the ways in which the be unsatisfying in others. The incommensurability
holon helps us accommodate this imperative. of contexts makes full resolution of conflicts unim-
aginable, i.e. every present solution is provisional,
and subsequent re-evaluations may arrive at a
Holons and change
different choice as the wisest. Further, the incom-
The holon exists, and seeks to persist, in its ecology mensurability of the holon’s contexts means that
of contexts. Individual contexts likely accommo- it cannot be fully optimized, for example, a success-
date, and indeed encourage, any of several alterna- ful farm can never be simply the diligent appli-
tive configurations of the holon. However, the cation of the contents of the collected technical
holon’s configuration must be at least tolerated in bulletins from the university.
all of its multiple, incommensurable contexts. For Every holon is unique in both its present state
us to even notice a holon in the first place it must and how it will react to changing contexts. The
have earlier found such a configuration. But an present state of the bundle of contexts in which it
instant later contexts and holons have changed – exists, and the path by which they evolved are
an illness or a price increase forces reassessment, unique, shaped by accidents of history. Thus a com-
and likely some reconfiguration, if the holon is to plete articulation of the present state of the holon is
persist. For better or for worse, change is inevita- impossible. Further, and perhaps more significantly
ble. This need to find a way of existing in an for the analyst, so too is predicting the holon’s reac-
ecology of contexts, and that this ecology forces tion to changing contexts. This unpredictability has
constant reconfiguration, motivates our contention multiple origins, including, at least, historical con-
(approaching insistence) that agroecosystem analy- tingency, un-unified intentionalities and the sense
sis should begin with a reverence for the farmer’s of permission to create that a holon may gain
organizational genius and planning. from its contexts, for example, democracy versus
The holonic interpretation we propose invites authoritarianism (Bell, forthcoming). Historical
and provides conceptual room for the change that contingency, that is, that what will happen here
is essential to carrying on in an ever-evolving and now, is powerfully shaped by the particular
environment – ‘the constant dance of cognitive history of the holon and, possibly, some of its
systems, continually shaping, learning, and adapt- important contexts. Legacies of past experience
ing to their environment . . .’ as Pretty (2002: 149) are embedded in the repertoire of reactions from
described a central idea of the biologist-philoso- which the holon draws as it reacts to a new environ-
phers Varela and Maturana. The concept of the ment – the schemata of Gell-Mann’s complex
holon introduces into the greater web of involve- adaptive systems (Gell-Mann, 1994, 1995).
ment an entity with intentionality, and this inten- The intentionality of the holon, in spite of its
tionality provides the motive force for the centrality, can never be unified – the holon’s inten-
ceaseless planning and action that is required to tionality is not singular, but rather always at odds
guide the change that is imperative to the holon’s with itself. There are two sources of disunity, the
survival. For Fuenmayor (1991) intentionality first of which are tensions that arise from the irre-
causes the holon to be ‘thrown’ from the present solvable task of seeking satisfaction simultaneously
towards a future state, but we envision a great in incommensurable contexts. Just as incommen-
deal more many-directional pushing and pulling, surability makes impossible the calculation of an
from and toward, as it moves through time, con- optimal outcome, it makes unimaginable a single,
tinuously reassessing its relational involvements. clear intentionality guiding the holon. The second
That the holon finds viable configurations does source of disjunct intentionalities is political –
not mean that it is free of internal tensions, for member holons of a holon may have different inter-
example, a grower of organic grains may feel ests and priorities. A farm holon that includes
uncomfortable with the repeated soil tillage members who have a passion for grassland birds
needed to manage weeds, given the implications and others who desire to be known as the owners
for erosion. While a particular solution may seem of the largest herd of cows in the state will likely
to conveniently and with little compromise be work from an un-unified intentionality. The rich
workable in two or three contexts, it likely will array of possibilities by which the holon might

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

292 W.L. BLAND AND M.M. BELL

react to changing contexts as a result of historical unable to agree on any distinctions in the gloom.
contingency and un-unified intentionalities should While we are free to choose whatever holonic
humble the analyst and policy maker. identifications we like, the analyst who exercises
But there is reason to believe that a holon’s beha- his or her intentionality with no regard to the
viour will not be random: intentionalities have holonic identifications of other intentionalities
directionalities from and towards that they and risks mistaking turnips for watches, wishes for
their (holonic) contexts try to maintain, and in horses, beggars for aristocrats. Holons may be con-
that sense have a kind of stability that is in some structions, but they have real consequences. Inten-
degree predictable. But at the same time, they tionalities may only imagine their holonic
carry on with those directionalities by constantly boundaries and their contexts, but the directions
reconfiguring their stabilities amid the unfinished their actions follow as a result shape their lives
possibilities of the world, in ways that neither and ours mightily. Intentional actions give persist-
they nor we can ever fully prejudge. This constant ence to flux, and flux to persistence. They create
seeing and reseeing of a holon’s boundaries is subjects and objects, texts and contexts.
another essential feature of the epistemology of The fluxing persistences and persisting fluxes
flickering. It entails an acceptance of directed wrought by intentionalities offer the analyst a great
unpredictability, and is a large part of what we conceptual opportunity: A really helpful way to
mean by holonic thinking’s reverence for planning. begin to understand the world is to try to understand
We thus stress the unfinished quality of holonic it as others do, and then watch carefully for the
involvement that we label unfinalizability, a term consequences of those understandings, with all
we borrow from Bakhtin (1981, 1984). There is their conflicts and asymmetries. Understand holons
always – always – slippage along, and within, a as holons understand themselves, and study as they
holonic boundary. There is always an untidiness try to organize and reorganize themselves and their
to reality. There is always tension between connect- contexts accordingly. Make it a crucial analytic act
edness and unconnectedness. And it is a good thing of the agroecological endeavour to look for holonic
too, for it is here in this potential for the continual boundaries that others are trying, with assuredly
reworking of that which was thought to be worked varying degrees of success, to draw, and to study
out that we discover the possibility for change, and the contextual relations that give such impetus.
indeed for life. A fully bounded and connected From the perspective of practical application,
world is a frozen world, incapable of supporting holon agroecology offers suggestions, if not a
life and its inherent capacity for change. Holonic rough framework, for students (be they under-
thinking, however, with its stereo, flickering view graduates or national officials) of agriculture as
is constantly in motion, never accepting an icy fina- they seek to understand or influence the behaviour
lization of involvement. As the holon continuously of those who seek livelihoods from plant and
rearranges itself, its external connections – its web animal increase. Because the farm holon must con-
of involvements – will also necessarily change. The stantly seek a configuration that is viable in mul-
holon perspective therefore accepts both an unpre- tiple, incommensurable contexts, the student
dictability of involvement and, fortunately, a should first seek to understand this constellation
certain predictability in that unpredictability. of contexts much as the farmer does. An openness
and humility is required to avoid overlooking
powerful, yet perhaps foreign (to the analyst) con-
Holons as narrative texts. Closely related is the narrative boundary
problem – farmer and agroecosystem analyst may
To engage in holonic thinking is, of course, to con- start from fundamentally different premises about
struct narratives. It is a way to tell stories about the the meanings of an agricultural endeavour.
world. It is not the world itself. (Nothing but the Further, the analyst is well served by delaying
world can make that claim.) But holonic thinking, normative judgments as long as possible, by
as we have been conceptualizing it, also offers an approaching the farm holon with a reverence for
argument for tracking through the endless swamp the organization and planning that is required to
of postmodern discourse analysis, in which we are create and maintain a farm holon.

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

 A HOLON APPROACH TO AGROECOLOGY 293

Conclusion which the recognition of intentionality is less directly

relevant to one’s purposes – when it is only, or nearly
The time is long past, or so we hope, when anyone only, the analyst’s own intentionality that shapes the
seriously considers understanding the involved huge- purpose of the encounter with the world – then a
ness of agroecology with the metaphor of the simple systems ladder is likely to be very useful.
machine. And yet the turn to systems thinking, as it But such a singularity is an ontological conceit,
is currently developed in much of agroecology, has and we forget this at our peril. Sustainability
moved on less than we often recognize. The simple entails contending with the disjunct openness of
systems approach, even when applied in a detailed an incomplete world of intentionalities that are
and elaborate way, retains many of the features of neither part nor whole. Sustainability means
the machine metaphor. Interactions are clear, comp- keeping life flickering amid these ongoing, if un-
lementary, balanced, and directed to common balanced and asymmetrical, consequences of unfina-
purpose in ways that yield the analyst a confident lizability. Sustainability, then, means possibility,
sense of interventionist power and control in a fully and possibility means intentionality.
connected world. Such a rhetoric and mood has a Koestler called the book in which he introduced
definite mechanistic feel, albeit assuredly a more the concept of the holon The Ghost in the Machine.
informed mechanism than the simple simple Our argument has been that intentionality is the
approach of the machine metaphor of earlier science. holonic ghost which prevents the world from ever
We have argued that the concept of the holon, in being a machine. It is this hopeful thought that
which neither parts nor wholes exist in an absolute we invite agroecologists to sustain.
sense, can be developed to appreciate the incomplete
and incommensurable quality of involvement that
stems from the unfinalizability of contextual inten-
Note
tionality. We have suggested tracing holon bound-
1. Although we do not list our names alphabetically,
aries by looking for the intentionalities that seek to the work (and responsibility) for this article is
construct themselves from the fluxing welter of equally shared by the two of us.
context. The agroecology analyst would do well to
take careful note of these constructions, their con-
ditions and their real consequences – an appreci- References
ation we have called a reverence for planning.
While we believe that there is reason to hope agroe- Ahl, V. and Allen, T.F.H. (1996) Hierarchy Theory: A
Vision, Vocabulary, and Epistemology. New York:
cologists will find our arguments worth engaging, we Columbia University Press.
worry that the influence of the language of systems Allen, T.F.H. and Starr, T.B. (1982) Hierarchy: Perspec-
thinking will make our argument seem oppositional, tives for Ecological Complexity. Chicago: University
and perhaps even threatening. Much has been learned of Chicago Press.
in agroecology through the application of a systems Allen, T.F.H., Tainter, J.A., Pires, J.C. and Hoekstra,
T.W. (2001) Dragnet Ecology – ‘Just the facts
point of view, and many may wonder why we ma’am’: The privilege of science in a postmodern
would appear to question that. We do not question world. BioScience 51, 475 –485.
that. Rather, we worry that we often reach beyond Aristotle (1987) A New Aristotle Reader. In J.L. Ackrill
the safe height of its conceptual ladder. Complexity (ed.). Princeton, NJ: Princeton University.
theory for some time has recognized that this limit Bakhtin, M. (1981) The Dialogic Imagination: Four
Essays. Austin, TX: University of Texas.
is often exceeded. Our goal has been to provide Bakhtin, M. (1984) Problems of Dostoevsky’s Poetics. In C.
more secure footings for the needed conceptual exten- Emerson (ed. and trans.) Introduction by Wayne C.
sions – the search for ‘a more comprehensive onto- Booth. Minneapolis, MN: University of Minnesota Press.
logical and epistemological framework for studying Bawden, R.J. and Ison, R.L. (1992) The purposes of field-
farm enterprises’, called for by Noe and Alrøe (2006). crop ecosystems: Social and economic aspects. In
C. Pearson (ed.), Field Crop Ecosystems (pp. 11–35).
We do not ask for the banishment of simple London: Elsevier.
systems thinking, then. There are times when a Bell, M.M. (forthcoming) Strange music: Notes toward a
shorter ladder is just what is needed. Situations in dialogic sociology. Humanity and Society.

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

294 W.L. BLAND AND M.M. BELL

Checkland, P. and Scholes, J. (1999) Soft Systems Meth- Noe, E. and Alrøe, H.F. (2006) Combining Luhmann
odology in Action, Including Soft Systems Method- and actor-network theory to see farm enterprises as
ology: A 30-Year Retrospective. New York: Wiley. self-organizing systems. Cybernetics and Human
Churchman, C.W. (1979) The Systems Approach and Its Knowing 13, 34–48.
Enemies. New York: Basic Books. Nussbaum, M.C. and Sen, A. (eds) (1993) The Quality of
Conway, G.R. (1987) The properties of agroecosystems. Life. Oxford: Clarendon Press.
Agricultural Systems 24, 95–117. Onions, C.T. (1955 (1933)) The Oxford Universal Dic-
Cronon, W. (1992) A place for stories: Nature, history tionary on Historical Principles. Oxford: Oxford at
and narrative. Journal of American History 78, the Clarendon Press.
1347–1376. Pearson, C. and Ison, R.L. (1987) Agronomy of
Emirbayer, M. and Mische, A. (1998) What is agency? Grassland Systems. Cambridge: Cambridge
American Journal of Sociology 103, 962 –1023. University Press.
Fuenmayor, R. (1991) The self-referential structure of an Peart, R.M. and Shoup, W.D. (2004) Agricultural
everyday-living situation: A phenomological ontology Systems Management: Optimizing Efficiency and Per-
for interpretive systemology. Systems Practice 4, formance. New York: Marcel Dekker.
449– 472. Pretty, J. (2002) Agri-Culture: Reconnecting People,
Gell-Mann, M. (1994) The Quark and the Jaguar. Land and Nature. London: Earthscan.
New York, NY: W.H. Freeman. Rosen, R. (1991) Life Itself: A Comprehensive Inquiry
Gell-Mann, M. (1995) Complex adaptive systems. In into the Nature, Origin, and Foundation of Life.
H. Morowitz and J. Singer (eds) The Mind, the New York: Columbia University Press.
Brain, and Complex Adaptive Systems (pp. 11–23). Rosen, R. (2000) Essays on Life Itself. New York:
New York, NY: Addison-Wesley. Columbia University Press.
Gliessman, S.R. (2004) Agroecology and agroecosys- Sen, A. (1992) Inequality Reexamined. Cambridge, MA:
tems. In D. Rickerl and C. Francis (eds) Agroecosystem Harvard University Press.
Analysis (pp. 19–29). Madison, WI: American Society Sen, A. (1999) Development as Freedom. New York:
of Agronomy. Knopf.
Koestler, A. (1967) The Ghost in the Machine. Spedding, C.R.W. (1988) An Introduction to Agricul-
New York: Macmillan. tural Systems. New York: Elsevier Applied Science.
Martin, J.L. (2003) What is field theory? American Ulrich, W. (1993) Some difficulties of ecological
Journal of Sociology 109, 1–49. thinking, considered from a critical systems perspec-
Midgly, G. (2000) Systemic Intervention: Philosophy, tive: A plea for critical holism. Systems Practice 6,
Methodology and Practice. New York: Kluwer 583–611.
Academic. Vayda, A.P. (1986) Holism and individualism in
Muir, J. (1911) My First Summer in the Sierra. Boston: ecological anthropology. Reviews in Anthropology
Houghton-Mifflin. 13, 295 –313.

INTERNATIONAL JOURNAL OF AGRICULTURAL SUSTAINABILITY 5(4) 2007, Pages 280–294

View publication stats