DUNCAN LEY Critic Marxism 1982

Structural Marxism and Human Geography: A Critical Assessment
Author(s): James Duncan and David Ley

Source: Annals of the Association of American Geographers, Vol. 72, No. 1 (Mar., 1982), pp.
30-59
Published by: Taylor & Francis, Ltd. on behalf of the Association of American Geographers
Stable URL: https://www.jstor.org/stable/2563223
Accessed: 04-10-2019 15:32 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Association of American Geographers, Taylor & Francis, Ltd. are collaborating with
JSTOR to digitize, preserve and extend access to Annals of the Association of American
Geographers
This content downloaded from 143.107.8.10 on Fri, 04 Oct 2019 15:32:56 UTC
All use subject to https://about.jstor.org/terms
Articles
Structural Marxism and Human Geography:

A Critical Assessment
James Duncan and David Ley
Department of Geography, University of British Columbia, Vancouver, British Columbia V6T 1 W5
Abstract. This paper assesses critically both the strong theoretical claims and the empirical work of a
number of structural marxists in analyzing the geography of advanced societies. Such work offers a
holistic mode of explanation that has important philosophical affinities with Hegelian idealism. In
explanation, reified entities such as capital are treated as the formal cause while people are regarded
as the efficient cause, the mere carriers of a structural logic. This perspective raises a number of serious
theoretical problems that are not resolved, including the status of individuals as a creative force in
shaping events, the ontological status of structures, the relationship between consciousness and struc-
ture, and the tendency to functionalism and teleology in explanation. These shortcomings have
severe consequences for empirical study. The fundamentally economic nature of the central
categories provide at best a partial analysis and not a general comprehension of society as a whole. If
applied literally, key categories, such as the dichotomous class system, fit poorly with empirical events.
However, if adjusted to match historical events, they depart markedly from the form of the theoretical
structure. The result is a confused interplay between theory and empirical study, and a tendency
toward the mystification of causal processes and the denigration of empirical study in order to sustain
the integrity of the theoretical argument.
Key Words: structuralism, functionalism, holism, reification, capital, class, agency, empirical study.
Anotable theoretical development in human 302, emphasis added). Later, he adds, "I re-
geography and related fields in sociology gard the channels through which surplus value
and planning over the past decade has been circulates as the arteries through which course all
the emergence of an ambitious radical tradition the relationships and interactions which define
derived from structural marxism. Aside from the totality of society" (Harvey 1973, p. 312,
some skirmishes in the literature, a critical as- emphasis added). So too, Richard Peet optimis-
sessment of this tradition has not yet been made tically identified a radical science that "strips
in geography, an oversight that is the more sur- away diversions, exposes existing explanations
prising in light of the extravagant claims that to criticism, provides alternative explanations
have often been advanced for the position which trace the 'relationship' between 'social
(see, however, Claval 1977 and Saey 1978). problems' at the surface and deep societal
David Harvey, perhaps the most widely read causes, and encourages people to engage in
marxist geographer, argues that "the only their own theory construction" (Peet 1977a). To
method capable of uniting disciplines in such a this ambitious manifesto is added the intoxicat-
fashion that they can grapple with issues such as ing promise of "a radical political program for
urbanization, economic development and the restructuring society," around which "develops a
environment, is that founded in a properly con- culture which reflects the experience and the
stituted version of dialectical materialism as it wishes of a re-awakened people" (Peet
operates within a structured totality in the sense 1 977a).
that Marx conceived of it" (Harvey 1973, p. Such confident and sweeping claims to meth-
30
Structural Marxism and Human Geography 31
odological, theoretical, and practical com- economics-though to suggest thot questions of

pleteness invite a critical response; indeed, we power necessarily imply a marxist perspective
might go further and state that any claim to uni- would be to ignore a range of alternative posi-
versalism, if it is to be taken seriously at all, re- tions in social theory. Marxion analysis has
quires the most searching and challenging helped to demystify the categories used to ex-
examination. But to undertake such a critique is amine market economies, revealing the social
daunting, not least for the range of what cur- interests and social relations of domination im-
rently passes for marxist analysis: a wide spec- plied by talken-for-granted economic categories.
trum of published work that makes the marxian However, as we shall show, marxian theory in
designation imprecise, if not self-contradictory geography is inclined to its own forms of mystifi-
(Hook 1980). Beneath this broad canopy there cation and conceptual and theoretical stasis.
are, however, two major traditions, which Urban geographers seeking to understand the
Gouldner (1980) has recently referred to as the industrial city have been commended to aban-
two marxisms. His distinction of scientific and don Robert Parlk's interpretation of Chicago, for
critical marxism has been widely recognized in the earlier analysis of Manchester by Engels
social theory, though a varying terminology has (Harvey 1973, pp. 131 -34). But in the preface
served to accentuate different aspects of the to The Condition of the Working Class in En-
dualism: for recent authors, the descriptive gland, the youthful Engels observed that "it is
couplets have also contrasted theoretical as op- essential to have an adequate knowledge of
posed to historical marxism, and structuralism as the way of life of the proletariat if one is to form
opposed to humanism (Thompson 1978b; a sound judgement on the merits of socialist
Johnson 1978). The distinction is frequently that theories.... Our German theorists have far too
of a theoretical treatment of political economy little knowledge of the world as it actually exists
vis-d-vis concrete historiographic study. For rea- . . . [they] held preconceived views about the
sons that need further clarification (though some proletariat, which are as ludicrous as they are
answers are suggested below), human geogra- absurd" (Engels 1958, pp. 3-4). It is ironic (as
phy has with surprisingly few exceptions appro- we shall argue) that much marxist worlk in ge-
priated the political economy tradition, which is ography has foundered upon exactly the some
conventionally presented as the orthodox shortcoming of conceptual and theoretical fore-
reading of Marx. In light of its dominance in closure that Engels criticized among his contem-
human geography, it will be the target of much poraries. The relations among theory, concept,
of our critique, though some attention will also and geographical circumstances emerge as
be given to the historiographic work as a highly problematic. This impasse is tied in part to
heterodoxy that adds its own challenges to the the abstract realms of Hegelian philosophy and
structural tradition. its holistic argument, where such terms as totality
Some might see this paper as an ungracious and essence square uncomfortably with empiri-
response to Peet's invitation "to see radical ge- cal analysis. It is precisely this philosophy of
ography systematically evaluated for what it has structure and holism that Harvey, Peet, and
to offer" (Peet 1978). The paper presents a others adhere to (as illustrated in the quotations
critique rather than an appreciation, for two above) which frustrates the empirical examina-
major reasons: firstly, the literature under discus- tion of advanced societies by marxist geog-
sion already makes its own case enthusiastically raphers. The problematic relations between
(but uncritically), and secondly, a final judgment philosophical holism and geographic reality will
would require far more space than is available provide the focus of our critique.
here. It is our expectation that critical analysis
permits more satisfactory scholarship to be at-
tained. Moreover, scepticism need not imply Marxism as Holistic Philosophy
outright rejection, and we would certainly sym-
pathize with some of the moral force of Marx's Holistic and Individualistic Explanation
condemnation of social relations in market
societies. So, too, it is through self-consciously Explanation in social science is often divided
marxian analysis in geography that discussions of into those explanations that are holistic and
power have been introduced, challenging the those that are nonholistic or individualistic, al-
politically inert categories of neoclassical though the latter can be a misleading term
32 Duncan and Ley
(Phillips 1976 Lukes 1973; O'Neill 1973; Ryan ducible to cumulative human actions and in-
1973). These are very broad categories, and teractions, and that the processes linking struc-
there are important differences of opinion ture or context and individual or social action
among the proponents of both forms of expla- need to be carefully specified.
nation. In general, social scientists who adopt Although any given individual may be rela-
the nonholist position believe that large-scale tively powerless in the face of a historic context,
social events such as war, culture, or capitalism these conditions are not active, transcendental
are the products of the interaction and interrela- forces beyond human activity. They have no
tions among the individuals who participate in telos, no inner logic, and no active causal prop-
these events, and that their explanation can in erties of their own. Though individuals may not
principle, although not necessarily in practice, be free to transcend their social context, neither
be reduced to statements about the actions of are they passive agents of a larger force such as
the participating individuals. Holists, on the other "culture," the "logic of capitalism," or the
hand, believe that such a reduction is theoreti- "mode of production."
cally impossible, holding that large-scale social Many of the marxists in geography draw freely
events are emergent and unrelated to the con- on a structural-holistic branch of marxism. Not all
scious actions of the individuals who participate marxists are holists, however; Marx himself at-
in them. Typically, a supra-individual entity sui tempted to reject the dualism between the indi-
generis, with a "logic" or properties funda- vidual and society and the ensuing need for a
mentally its own, is portrayed as the active transcendental object. In some writings, how-
causal variable, while individuals are viewed as ever, he tended to lapse into a holistic interpre-
passive agents of this larger entity. There is an tation of economic structures. This tension be-
essential ontological difference between these tween individualistic and holistic explanations
two positions. It is not merely a matter of how was inherited by Lenin, who was never able to
much freedom particular people may have in resolve it. Althusser, and many marxists since
the face of their sociohistorical context, which of Marx, have resolved it in favor of holism.
course is an empirical question; nor is it a ques- Of the different varieties of holism, the one
tion of the scale at which one chooses to view a that principally concerns us here is known as or-
particular social phenomenon. ganicism, and contains the major claims of most
Although in this paper we will argue against a holists in social science.1 The five basic claims of
holistic form of explanation that can be found in organicism are as follows:
the writing of many structural marxist geog- (1) The analytic approach proves inadequate
raphers, this is not to endorse a position that is when applied to society or to reality as a whole.
individualistic or idealistic in the-sense that direct (2) The whole is more than the sum of the
reference only to ideas, reasons, or intentions is parts. From a knowledge of the parts alone, the
likely to be adequate for any given explana- emergent properties of the whole cannot be
tion. Quite to the contrary, in many areas of predicted. After knowledge of the whole has
research the focus of attention is properly on somehow been obtained, this cannot be re-
macrostructures, whereas individual action can- duced to the properties and summation of the
not be fully explained without reference to the parts.
contexts under which individuals act. Reasons (3) The whole determines the nature of the
need not be synonymous with causes; much of parts.
social life is a result of the unintended conse- (4) The parts cannot be understood in isola-
quences of the cumulative actions of large tion from the whole.
numbers of past and present actors. The relations (5) The parts are dynamically interrelated
among the many actors may become pat- and interdependent (Phillips 1976, p. 6).
terned or structured; they may become in- Organicism is very different from the claim that it
stitutionalized and taken for granted. Individuals is sometimes useful to refer to entities such as
may be unaware of all the causes of their be- institutions rather than naming all of the indi-
havior, and intentions may form only a portion viduals that compose the institution. It is also
of the explanation of action. However, we different from the claim that although institutions
suggest that macroscale social structures should have an ontological status they are humanly
be precisely defined, that they do not have au- created and maintained and have no self-
tonomy or an existence that is not ultimately re- determination (Phillips 1976, p. 6). As we shall
see, the position of many marxists in geography vidual, one takes a world historical perspective,
is much closer to what we have identified as then one can see that there is "an inner logos to
organicism. It is understandable that it should be the seemingly chaotic multiplicity of events"
so, given the Hegelian background out of which (Bernstein 1971, p. 18). Although the history of
marxism arose. the world may appear to be the antithesis of
rationality, Hegel argued that this apparent
confusion is merely "the cunning of Reason,"
The Holism of Hegel which, through conflict, drives Geist toward its
own rational ends. Hegel placed great stress
It has been argued that Hegel's idealist notion upon viewing the world holistically, for its es-
of Geist, or world spirit, has served as a pro- sence is Geist, which pervades everything.
totype for Durkheim's "collective consciousness" Geist is the driving force behind specific peoples
and Kroeber's superorganic concept of culture and civilizations, evolving dialectically through
(Duncan 1980). Hegel's concept of Geist has conflict and tension between opposing forces
also greatly influenced marxists and through (Hegel 1953, p. 89). It is this dialectic of con-
them marxist geographers, and therefore it will flicting forces that accounts for movement in
repay us to examine his ideas more closely. For history; it is, to use the colorful phrase favored by
Hegel, true knowledge lies in understanding some holists, "the motor of history."
Geist and its relationship to the world. Hegel In short, Hegel viewed human beings as es-
considered Geist to be all-comprehensive and sentially passive, as incapable of shaping the
active, a force that, in the course of its develop- world. It is Geist that is the subject and men,
ment, has divided itself into a realm of spirit and women, and the objects of the world that are
a realm of nature. He argued, however, that the predicate. Hegel did not view Geist as sep-
this division of Geist into natural and spiritual arate and isolated from the world of human
spheres is only temporary and that eventually it beings and objects but as internally related with
will be aufgehoben, that is, it will transcend itself it, as different sides of the same thing, that is, he
and become one. Geist is not something apart saw them dialectically. Each needs the other to
from nature, for nature is a part of Geist. Hegel affirm itself, each is a facet of the other, moving
did not deny the ontological status of objects in toward the other, toward unity, oneness.
the world; he merely claimed that Geist is act-
ing in and through these objects. In order to un-
derstand Geist, then, "we must understand its From Hegel to Marx
concrete working in the 'material' of the world"
(Bernstein 1971, p. 30). Geist must actualize Marx came under the sway of Hegelian
itself, which it does by objectifying itself in the philosophy early in his career, and although he
world. All objects in the world, although they subsequently broke violently with Hegelianism,
may appear as nothing other than brute objects he did not totally escape its influence. Hegel's
to the ignorant, are in fact Geist. The implica- impact upon Marx is common knowledge; in
tions of this are immense, for this means that fact, according to Dupr6 (1974, p. 703), "If
Geist is both mountains and institutions, oceans there is anything that characterizes present day
and men's passions-it is all things. Of all the Marxist scholarship in the West, it is its concern
objects in the world, however, it is man who with the relationship of the ideas of the great
plays the central role in the development of German idealist to those of Marx."2 The perva-
Geist. It is man, according to Hegel, who is the sive infiltration of Hegelian influences in Marx's
agent of Geist and through whom Geist ex- writing has been noted by critics like Kolakowski
presses itself in history. Hegel relied on the aris- (1978), revisionists like Althusser (1969) (who
totelian distinction between efficient and formal regard them as pollutants to be expunged),
causes; Geist is the formal cause and man is the and by more sympathetic authors such as
efficient cause through which Geist functions. Bernstein (1971) and Oilman (1971). Ollman's
From the notion that Geist, which also can be (1971) discussion is particularly relevant, for it
translated as Reason, guides the world, it is but a has provided an important source for radical
short step to the position that the world is ra- geographers (see, for example, Harvey 1973,
tional. Hegel claimed that if, rather than looking pp. 287-89). He develops at some length the
at the world from the viewpoint of the indi- logical continuities between Hegel and Marx:
34 Duncan and Ley
Hegel constructed the framework described here in tained a teleological perspective on history.
order to treat ideas, characterizing what I have Rather than the Hegelian view of Geist gradu-
called the "whole" as "Absolute Idea" or "Reason."
ally transforming itself as it moves toward a
Marx's criticism is always directed against how Hegel
chose to apply this framework and his preferred higher and higher plane on its quest for the in-
subject matter, and never against the relational evitable perfection of world unity, Marx's trans-
quality of his units or the fact of system which this formation yielded a view of human beings
entails (Ollman 1971, p. 34).
gradually transforming themselves as they
move inexorably upward from stage to stage
Marx broke with Hegel over the issue of ide-
until they reach social perfection in communist
alism. To Marx it was not Geist that needed to
society. Similarly, Marx followed Hegel in view-
objectify itself, it was people who had this need.
ing the process by which history unfolds as the
Marx arrived at this position through his reading
dialectical process of conflict between opposing
of Feuerbach's critique of Hegel. In his book, The
forces which yields a new synthesis which is in
Essence of Christianity, Feuerbach (1957) in
turn contradicted, and so on, onward and up-
effect turned Hegel's philosophy "upside down."
ward. Again, it has been argued that Marx, like
He did this by subjecting Hegel's philosophy to
Hegel, subscribed to a philosophy of internal re-
what he called transformational criticism, which
lations (Oilman 1971, p. 37). In this view all
simply consists of inverting the subject and the
things are internally related to each other. "The
predicate. Geist is no longer the subject, the
full truth about any one thing includes (because
formal cause, which expresses itself through in-
of internal relations) the truth about everything"
dividuals, the efficient cause. For Feuerbach,
(Oilman 1971, p. 38). The whole is not simply
people are the subject, the formal cause, and
conceived of as the sum of the parts, for the
Geist is simply a human product, a product of
whole is an element of each part. Knowledge
alienation. Marx clearly showed the influence of
of a part therefore presupposes a knowledge of
Feuerbach's ideas in his early as well as his later
the whole.
work, although he was critical of Feuerbach for
not going far enough in eliminating idealism
from his concept of human beings (Marx 1 967b,
pp. 316-17; Marx 1967a, Vol. 1, p. 19; Marx A Structural Marxist Tradition
1972, p. 107; Schrag 1975, p. 25). However,
despite Marx's opposition to idealism, he in
Many marxist geographers have apparently
places lapsed into a transcendental idealist
been influenced by such writers as Althusser,
mode of thinking himself, seemingly endowing
Castells, and the French School of structural marx-
such entities as "capital" with an inexorable and
ism as well as by Marx himself. Althusser and
mysterious power over men and women.
the French structural marxists have taken the
Thompson argues this point and claims that
Hegelian idealist elements from Marx and de-
"capital has become an idea which unfolds itself
veloped them to an extreme form. This is de-
in history." He continues:
spite the fact that they claim to despise Hegelian
marxism, by which they refer to the work of
In the Grundrisse-and not once or twice, but in
the whole mode of presentation-we have exam- certain marxists who emphasize the social-
ples of unreconstructed Hegelianism. Capital psychological concept of alienation also derived
posits conditions "in accordance with its immanent from Hegel. Althusser believes that Marx even-
essence." . . . Capital posits this and that, creates tually rejected Hegelian idealism in part be-
this and that, and if we are to conceive of
cause Geist, the "Idea," is replaced by struc-
capitalism ("the inner construction of modern soci-
ety") it can only be as "capital in the totality of its tures. However, Althusser himself states that he
relations" (Thompson 1978b, p. 253).3 has retained the priority of the totality over the
elements, the dialectic, contradictions, and also
Although Marx's transformation of Hegel re- the belief that human beings are merely "bear-
moved the concept of Geist, the transformation ers" or agents of structures (Althusser 1972, p.
retained much of the basic structure of the 183).4 One could argue that structures, because
Hegelian argument. The world spirit was re- they are above and beyond the empirical, con-
placed in Marx by a world system, the system of crete world, are equally as transcendental as
capitalism (Hamilton 1974, p. 90). He also re- Geist; the mode of reasoning, therefore, is pure
idealism. Other examples of the Hegelion portant elements of Oilman's interpretation of

overtones of Althusser's writings include his indif- Marx. In the final chapter of Social Justice and
ference to an empirical verification of theories the City, he lays out an ontology, which is that of
and his comparison of true knowledge to math- internal relations (Harvey 1973, p. 288). He ar-
ematics whose truth is internal to its logic, not gues, following Oilman, that society must be
dependent upon external reality (Althusser and understood as a "totality of internally related
Balibar 1970, p. 59). In fact, as E. P. Thompson, parts with inner laws of transformation" (Harvey
a marxist who rejects holism, has declared: 1973, p. 296). Anderson (1973, p. 3) writes
that "a port . .. must be . .. situated within the
If there is a "marxism" of the contemporary world totality that gives it its specific significance," while
which Marx or Engels would have recognized in- in order to understand facts one must under-
stantly as an idealism, Althusserian structuralism is
stand their "relations to their own past develop-
this. The category has attained a primacy over its
material referent; the conceptual structure hangs ment and inherent potentialities for the future."
above and dominates social being (Thompson Santos (1977, p. 4) claims that one must grasp
1978b, p. 205). "the particular as a scission of the whole, a mo-
ment of the whole as well as the whole repro-
Contemporary structural marxism can be duced in one of its fractions." So too Walker
subjected to Marx's own critique of reification, by (1978, p. 168) states that:
which he objected to "seeing relations between
men as relations between things" and to Capitalist urbanization partakes of the contradictions
and movements of capitalist development. Urbani-
analyses that distort social reality by "obscuring
zation is not, of course, a mere reflection of an or-
the latter's character as an ongoing human pro- ganic totality, but rather an internal relation of the
duction, viewing it instead in thing-like whole of capitalism.
categories appropriate only to the world of na-
ture (Marx 1 967b, p. 425; Berger and Luckman Elsewhere we find that the socioeconomic for-
1967, p. 61; Israel 1971). The critique can also mation, in typically Hegelian fashion, "expresses
be applied to Marx's own work in places, as has the unity and the totality of the diverse
been pointed out earlier. spheres-economic, social, political, cultural"
While we do not mean to imply that marxist (Santos 1977, p. 4). Harvey refers frequently to
geographers are all to be associated with Althus- a "totality" being "structured by the elaboration
ser's extreme version of structuralism, although of relationships within it" and a totality such as
some clearly are, we intend to show that their capitalism that "seeks to shape the parts so that
work represents a continuity with this structural each part functions to preserve the existence
tradition and not a radical break from it (Peet and general structure of the whole." He con-
and Lyons 1981).5 tinues, "transformations" of labor power, surplus,
or other "reflections of all social relationships
within a given mode of production" occur
through contradictions, thus restructuring the
The Holist Claims of Marxist Geography
whole. Thus, "the evolution of society as a total-
ity must therefore be interpreted as the result of
Many of the marxist geographers explicitly contradictions established both within and be-
support the holistic claims of structural marxism, tween structures" (Harvey 1973, pp. 288-89,
and as such they are linked to the idealist Hegel- 293). Furthermore, the meaning of a society
ian background that we have developed "can be grasped only on the level of the totality"
above.6 Capitalism, Society, or other hypos- (Santos 1977, p. 7).
tatized abstractions are substituted for Hegel's
Geist. One can see that the language in the
following quotations from the work of geog- Marxism as Reified Theory
raphers represents Hegelian ontology in which
parts are "internally related manifestations" or The Geographers' Use of Reified Entities
"moments" in the development of the whole.
David Harvey, perhaps the most influential of Holism usually involves reification, which is a
the marxists in geography, draws heavily on im- fallacy by which mental constructions or abstrac-
36 Duncan and Ley
tions are seen as having substance and causal that "gather speed" (Santos 1977, pp. 4-5;
efficacy (Berger and Pullberg 1964-65; Israel Santos 1975, pp. 353-54). Peet, following Al-
1971; Keat and Urry 1975). These abstractions thusser, writes: "Each instance of the social for-
are usually portrayed as organismic or machine- mation moves through time with its own rhythm,
like entities having self-direction and power over unevenly developing relative to other instances,
women and men. Examples of such reified with which it nevertheless is interrelated into an
supra-individual entities, found in the writings of organic whole" (Peet and Lyons 1981, p. 194).
the marxist geographers, include: capitalism, Blaut (1975, pp. 3, 6) also seems to favor the
capital, labor, social formation, capitalist mode use of organic terminology, speaking of nations
of production, contradictions, the state, class, as hybrid organisms, and capitalism as an
society, and the market. Despite the fact that in evolving organism with survival strategies. Marx-
various programmatic statements structures are ist geographers have not been alone in using
said to be dialectically related to individuals, it is organic analogies. Most holists, from Hegel on-
supra-individual wholes that are invariably the ward, have believed that supra-individual
active subjects in the marxist geographers' anal- wholes are greater than the sums of their indi-
ysis while individuals, the parts, are the objects vidual parts and that the biological organism is a
acted upon. Castells and Harvey in the following prime example of a whole that is greater than
two statements assert that the terms class and its parts.8 However, few biologists today would
capitalism are not just summary descriptions for subscribe to that theory, and, furthermore, the
capitalists, so that one cannot interpret words processes being compared are fundamentally
such as capital as simply shorthand terms mean- different.9 Evolutionary terminology has in the
ing groups of capitalists. For example, Castells post been common in several of the social sci-
(1977, p. 320) writes: "But if the representatives ences, including cultural geography and
of a social class do not always know how to rec- anthropology. Such language is usually indica-
ognize themselves, the class itself knows its own tive of teleological or functional reasoning.
interests, in the sense that its unconscious logic Phenomena evolve in particular ways because
tends to sweep away whatever does not serve these are useful or essential to their survival. As
its interests."7 Here class is presented as an entity we will explain below, such statements tend to
that acts in spite of the inaction of members of be tautological and empirically untestable,
the class, who are passive agents swept along by therefore useless as forms of explanation.
its "unconscious logic." Harvey (1975b, p. 20) Some of the other reifications commonly used
writes that capitalism "opens up fresh pastures for in this literature evoke mechanistic and physical
the bourgeoisie to accomplish its historic mission." metaphors. Examples include: the mode of
Here, capitalism works in the interest of the production drives the social formation through
bourgeoisie but clearly is distinguishable from time, contradictions fly off and become em-
members of the bourgeoisie. Often the reifica- bedded, contradictions erupt and propel
tions take on an organismic quality, for exam- capitalism, contradictions force their way to the
ple, capital "needing" to increase and multiply, surface, and imperialism is the engine which
capitalism maturing, and imperialism maturing powers the system of capitalism (Peet 1979, p.
(Santos 1975, p. 347; Blaut 1975, pp. 6, 13). 167; Peet 1977b; Walker 1978, p. 169; Blaut
Or again, there are such phrases as the mode of 1975, p. 13). One wonders just what could pos-
production makes demands, capitalism devises sibly be meant by such highly reified and
solutions, capital throws its weight, social forma- anthropomorphic statements as the following:
tions and modes of production write history, the
market mechanism is the culprit, and history vic-
timizes people (Walker 1978, pp. 168-69; As class relations move over space, they pick up
qualities from the regions of a social formation, that
Harvey 1978a, p. 14; Santos 1977, p. 5; Har-
is, from the history of the local interaction between
vey 1972, p. 10; Harvey 1975d, p. 54). Santos a mode of production and environmental space,
speaks of the socioeconomic formation as an and at a wider scale from the histories of other social
organism, society as a "coherent social or- formations. Class relations become infused with the
direct and indirect effects of the contents of regions
ganism" that evolves according to "systematic
and environments, previous and "future" moments
laws," and the mode of production as a
in the history of a mode of production (which have
"genus," the social formation as a "species." He become temporarily frozen in space) ... (Peet
refers to rising productive forces as "mutations" 1979, p. 167).
The subjects of these sentences are so abstracted that to the extent that marxist geographers be-
from reality that the sentences become procti- lieve capitalism to be a determining force, other
colly meaningless. Some of the statements listed types of explanation are believed unnecessary,
above may be metaphorical; however, al- and many important explanatory factors are
though they may be intended to be evocative precluded from the analysis.
or polemical, they nevertheless must be crit-
icized as irresponsible. One does not always
know which metaphors are to be taken seriously
The Logic of Wholes
as explanations and which are not. There is a
fundamental mystification concerning process.
We do not know what "capital throwing its The reified categories of the marxists are not
weight" means; statements such as this do not only granted a life of their own, but they also
constitute scientific analysis. However, we do have a purpose, or a telos: they move toward
know that capital is used here in a reified sense some historically determined end. Clearly these
and that whatever processes are at work are wholes are not guided by the will of people (the
being obscured. Other statements, however, parts) for, as we shall discover, this is ruled out.
are perhaps meant to be taken seriously. How, then, are the wholes guided down "the
Statements such as "capitalism devises solutions" path of history?" The answer is that the wholes
or "the mode of production drives the social are thought to have a set of internal laws, an
formation through time" seem to be offered as inner logic. Santos (1977, p. 4) expresses this
analytic statements. Again, however, because belief when he writes:
capitalism or mode of production are not
"things" that can seek or intend, the true under- "Stages along a process" . . socioeconomic for-
mations can be understood only within the
lying processes at work are mystified.
framework of a totalizing movement, all of whose
Reification, which has sometimes been re- elements are variables interacting and evolving to-
ferred to as the fallacy of misplaced concrete- gether, governed by the law of the whole. Society
ness, is an empirical rather than a logical error. It evolves systematically, as "a coherent social or-
ganism whose systematic laws .. . were the su-
involves attributing substance, power, activity,
preme laws, and standard measures for all the
and sometimes intentionality to abstract con-
other, more specific regularities."
cepts. One common claim of holism, that the
whole determines the nature of the parts, is This claim is of course in line with the holist view
found in the writing of the marxist geographers that structures must be analyzed at their own
in the form of capital or some other reified entity level, that they are not reducible to a lower
causing men to act in certain ways regardless of level. But we are reminded to sift carefully
their reasons, intentions, or desires. This type of through the vast number of events that make up
analysis obscures the true underlying processes the world in order to discern this logic. Some
by which the actions of individuals produce the facts are critical to "the logic" while others are
structural conditions under which they act and merely "noise." In this regard:
through which these conditions may later be-
come constraining or coercive (Agnew and "Special factors" also arise in relation to foreign
Duncan 1981; Lukes 1977). Although the broad trade which can confuse, conceal and distort mat-
social and economic context within which indi- ters. The significance of such factors to actual histori-
cal situations is not denied . . . they are just not re-
viduals act may not be directly traceable to
garded as crucial for understanding the inner logic
specifiable actors, to suggest that it must, then, of the capitalist mode of production (Harvey
be determined by forces above and beyond 1975b, p. 14).
men and women is a mystification. Unintended
consequences of action may at times be in di- Upon reading this quotation, one is immediately
rect contradiction to the intentions of individuals reminded of Hegel's views on an inner logic.
who act. However, although suprahuman forces Hegel also believed that there is an "inner logic"
such as capitalism appear autonomous, it may at work "driving history" inexorably forward to its
simply be because the multitude of actions both preordained end. Hegel, like marxist geog-
historical and contemporary that produce raphers, believed that this inner logic is difficult to
capitalist structures appear anonymous. Fur- discern, that one has to sift through a "seemingly
thermore, and most importantly, one could say chaotic multiplicity of events" in order to grasp it.
38 Duncan and Ley
It is held that some observers may occasionally these forces are. Because these forces collec-
catch glimpses of this logic, sufficient at least to tively determine "outcomes" it is an issue of no
allow them to move forward to a higher level of small importance. What, for example, is the
awareness, that is, to move history. Marxists, in difference between "economic" and "competi-
contrast, are held to be better equipped to de- tive" forces, between "political," "ideological,"
tect this logic: and "legal" forces, and how are they all distinct
from what are identified as "social" forces? And
We will usually be able to identify the underlying what might the category labeled "and the like"
logic dictated by capital accumulation at work. But
contain? If our understanding of the logic entails
the underlying logic does not, and indeed cannot
uniquely determine outcomes. The latter have to an understanding of these "forces" as Harvey
be understood in terms of the balance of forces ... describes them, then our understanding will in-
economic, social, political, ideological, competi- deed remain abstract. If we are told that out-
tive, legal, military, and the like . .. through which comes are a product of a hidden logic plus
interest groups and classes become conscious of the
everything else, then for from providing a pro-
contradictory underlying logic and seek by their ac-
tions to "fight it out" to some sort of resolution (Har- found insight, one has not left the prescientific
vey 1975b, p. 18). realms of tautology and nominalism.
This is an interesting statement in part because it

is so Hegelian. Just as Hegel tells us that when
People as Agents of Wholes
people become conscious of the underlying
logic of the world, when they realize that Geist
is everything, then they will arrive at the resolu- We have seen that structural marxists reify en-
tion of history; so we are told here that a resolu- tities such as capital and the mode of production
tion of sorts is to be achieved when people and endow them with a power to do things. It is
become conscious of the underlying logic. Fur- these reified wholes which shape the course of
thermore, this statement raises the issue of what history. The "guide" or "pattern" that determines
this internal logic or low is; presumably it refers to the direction of these wholes is the "internal
the manner in which some process operates. laws" or the "logic" of the whole. It is this which
Harvey claims that marxists can understand this shapes the course of these wholes as they move
process, that they can decipher the logic. How- toward their ultimate goal-communism. The
ever, on the some page we are told that the question arises, how do these wholes actually
task for geographers is "to specify how the 'inner get translated into the concrete contents of the
logic' of the capitalist mode of production, world? How, for example, does capital manage
abstractly conceived, relates to the concrete to do things, how does it conduct its "historic mis-
realities, the phenomenal forms of the historical sion?" The answer is that it accomplishes its tasks
process" (Harvey 1975b, p. 18, emphasis through women and men who are conceived of
added). In other words, geographers under- as its agents. Here as in Hegel we find the aris-
stand the logic only in the abstract but not as it totelian distinction between formal and efficient
relates to real circumstances. This is a very real cause. The reified entity, capital, for example, is
problem if one hopes to translate the insight into the formal, i.e., true, cause and human beings
practical action in the everyday world. Further- are simply the efficient cause. In other words,
more, there is the statement that the underlying people are not really a true cause at all, they
logic of capital accumulation does not itself de- are simply the means of implementing the
termine outcomes, that we must understand it in goals of the formal cause. This distinction is a
terms of a balance of forces, "economic, social, favorite device of those who wish to posit a
political, ideological, competitive, legal, mili- supra-individual entity while at the some time
tary, and the like." This merely raises a new set somehow account for the fact that people do in
of problems. First of all, we are told that the logic fact appear to do things that produce results.
applies to capital accumulation and this must be Hegel used this device to explain how men and
balanced against other "forces." Do these forces women were simply the agents of Geist, while
each have their own logic? Presumably they do, Durkheim used it to show that they were the
although nowhere is this spelled out. More agents of "social facts," and cultural anthropolo-
troubling than this is the issue of what exactly gists and cultural geographers used it to show
how individuals were the agents of a superor- (1974, p. 39) speaks of the movement of Third
ganic culture (Duncan 1980). World people to Americon ghettos: "I prefer
Among the structural marxists, Althusser is most 'tronslocotion' to 'migration' because individual
explicit on this subject (Aithusser and Balibar choice has no theoretical relevance here. (The
1970, p. 180). However, Preteceille (1976) 'decision to migrate,' in this case, is rather like
also speaks of the "different social agents" of the the decision to run from a burning building. So
State. People are merely agents of capitalism, much for migration theory.)" And so much for
dupes for whom, to quote Anderson (1973, p. the perceptions and decision making of people
2), "capitalism necessarily appears ... as in the Third World! Though Blaut rightly criticizes
something other than it really is." People are not the individualism of migration theory for positing
conditionally naive but are portrayed as neces- an individual who makes decisions "all alone"
sarily so, for if the reified entities are to be truly outside any social context, to offer in its stead the
active, then individuals must be conceived of as holist alternative, the structural model where the
passive and ignorant. Harvey adopts a similar individual is reduced to a "nondecisionmaker" is
stance when he speaks of "a faction of capital equally one sided. The literature is rife with sim-
seeking the appropriation of rent either directly ilar examples of the rejection of reasons in favor
(as landlords, property companies, and so on) of the "Reason" inherent in a transcendental
or indirectly (as financial intermediaries) system. Piclkvance (1976a, p. 65), for example,
Here again we find individuals as the agents of points out that Preteceille, in his study of the
the formal cause, landlords and other minions grands ensembles,
constantly ready to carry out the tasks assigned
to them by capital. But capital has other agents deliberately eschews questions such as whether the
residents are "satisfied," . . . Similarly, the concep-
as well. It usually "intervenes" in the built envi-
tion of buildings as an expression of the architect's
ronment "through the agency of state power"
intentions, or the more "sociological" conception
(Harvey 1978a, pp. 9, 14). The state also has its which sees them as the outcome of the strategies of
agents, which suggests that even the concept of various "actors," are both rejected. Such concep-
human agents as the efficient cause retreats tions give priorities to the actor(s) rather than to the
conditions under which they act.
farther into the background. In fact, at times
human agency seems to disappear altogether
and we are left with sets of interlocking abstrac- The conditions under which people act are the
tions, a world seen as a set of abstract categories real causes, according to these theorists. Appar-
filled by roles. An example of this is Harvey's ently none can escape the power of this system,
(1 978a, p. 12) statement that "capital will likely not even the most powerful government offi-
flow into landlordism." One is reminded here of cials, for "what is involved is the compulsion of
a comment made by E. P. Thompson (1 978b, a system in a rapidly advancing state of con-
p. 267) about the degree of reification in the tradiction. In such a situation the 'good inten-
work of the structural functionalist sociologists, tions' of this or that president count for very little"
who are, as we shall see later, intellectual bed- (Peet 1 977b). Not even the will of the capitalist
fellows of the marxist geographers. "Systems seems to make any difference. Harvey (1 975b,
and subsystems," he writes, "elements and p. 9) is quite blunt about this when he writes with
structures, are drilled up and down the pages reference to Marx's view of accumulation as a
pretending to be people." need of capitalism: "Yet this historical mission
does not stem from the inherent greed of the
capitalist; it arises, rather, out of forces entirely
independent of the capitalist's individual will"
The Relationship between Consciousness
(emphasis added). It is inconsistent, however, to
and Structure
describe these forces as "entirely independent,"
for earlier we have seen that the author accepts
There is a great deal of confusion in this litera- the notion of internal relations, the whole point
ture surrounding the relationship between con- of which is that there is no such thing as an entity
sciousness and structures. In much of the writing, that is "entirely independent" of any other.
the position appears to be that consciousness is Again we do not take issue with the notion
of no consequence in explanation. Thus Blaut that the structural conditions under which people
40 Duncan and Ley
act are critical to understanding action. How- "the drive on the part of capitol to inculcate the
ever, to suggest that these conditions are so to- working class with the 'work ethic' and
tally separate from people's beliefs, intentions, 'bourgeois values"' (Harvey 1978a, p. 22). Is
and strategies that one may ignore conscious- one to assume that it is not the capitalists who
ness is to engage in overstatement. Fortunately, are knowingly trying to instill these values in the
one can retain a concept of people's being af- working class but, rather, it is the system that is
fected by and in turn affecting structures without doing it? Gray (1976, pp. 40-41 ) sums up the
positing "forces entirely independent of ... structural marxist view on this issue as follows:
[people's] wills." In short, it is not the presence of
constraint that is at fault, but the reification of Now each of these classes, in order to continue the
structures and the resultant truncated model of existence of the social formation, must be repro-
duced and given a set of ideas, beliefs and as-
human beings in this work.
sumptions fitting their position in society. In other
Accounts of the process by which the "compul-
words, each class according to its position in the divi-
sion of the system" actually forces individuals to sion of labor must be socialized and receive a different
carry out the system's will are very sketchy and ideology suitable to its class position and destiny.
unsatisfactory. Harvey (1978a, p. 22), discussing
Here we can clearly see the imperatives of the
how the system translates its needs into behavior
system, that each class is given a set of ideas tai-
and consciousness, writes, "the new economic
lored to its position in the whole. And who or
order also required that [individuals] . . . had to
what does the giving? It would seem to be some
be made obedient to the cash stimulus and
supra-individual entity, which is trying to repro-
obedient in such a way as to react precisely to
duce itself by giving people ideas that are useful
the stimuli provided." Perhaps here we are be-
to its survival. Anderson (1973, pp. 2, 3), who
ginning to get at some of the mechanisms in-
appears to subscribe to this view, writes: "The
volved. Peet (1975, pp. 567-68) suggests
division of knowledge is useful to the rulers as a
further that marxist geographers should engage
control mechanism." One might think from this
in micro-scale analysis that deals with the "com-
statement that the "rulers" are consciously trying
plex of forces, both stimuli and frictions, which
to manipulate knowledge in order to control
immediately shape the course of a person's
people. But this turns out not to be so, for
life." The mechanisms being suggested here are
Anderson tells us that ideology is not to be seen
described in terms of a stimulus-response
in an "idealist" sense as a conspiracy. Again, it
model, whereby behavior is explained in terms
apparently is the system that is the subject,
of positive stimuli such as cash and negative
working toward its own functional ends through
stimuli or frictions. A stimulus-response model. is
people who are its unconscious agents. In a sim-
particularly well suited to theorists who posit a
ilar vein Short (1976, p. 80) argues that:
transcendental entity that causes behavior be-
cause the entity is conceived of as the subject
To understand the impact of ideology it is necessary
(stimulus) and individuals as objects that re-
to look at its function in capitalism. Its main function
spond. The model is attractive to holists because is to suggest to people that their actions spring from
it "black boxes" mind and hence no discussion of their own free will rather than as an adaptation to
individual consciousness or intersubjectivity is the economic system.
necessary beyond establishing that people re-

spond to the stimuli. This ideology "does not give individuals a true
The stimulus-response model is usually com- knowledge of the social structure but inserts
bined with a model of habituation, whereby them into their practical activities supporting this
people become conditioned to a stimulus and structure" (Short 1976, pp. 80-81 ). On reading
automatically, unthinkingly, respond to it. The these lines, one is struck by a sense of dojf vu,
term most often used when referring to habitua- for was it not Hegel who argued that people in
tion is "ideology." The structure, it is held, instills their ignorance believed that they had free will
in individuals a particular habitual way of think- and that they failed to see that man simply "re-
ing in order that they might carry out the will of flects the development of Geist and is the agent
the structure. Each class is given a type of con- through which Geist expresses itself in history"
sciousness or ideology which individuals appear (Bernstein 1971, pp. 21-22). The words have
to internalize en masse. In this regard, there is changed, of course, for structural marxists reject
the idea that action is directed by a reified human, and how have both of these collective
supra-individual entity termed a world spirit. Ac- characteristics survived the crushing determinism
tion for them is directed by "capitalism" or the of the transcendental structures? We are not told.
"social formation." But, whereas the terms may There is here an unwillingness to deal with the
be different, the type of explanation is the admittedly difficult questions of the empirical
same, for in both cases we find a supra- reality of relations between consciousness and
individual entity supposedly directing action. structure. All of these problems stem from the
Marxist geographers appear to support what ontological status attributed to structures.
Dennis Wrong (1961) has termed an over-
socialized view of man. Each class internalizes
the views and patterns of conduct imposed
Functionalism and Teleology
upon it by the reified entity in order for that entity
to fulfill its "historic mission." People are there-
fore viewed as varying only according to their Functionalism has a long and controversial
assignment in a social class, and behaving and history in social science. Modern functionalism
thinking in a standardized, habitual fashion may be traced back to the work of such
within the class. There seems to be no need for a nineteenth-century theorists as Darwin, Marx,
social psychology when one's view of human and Durkheim, and in the twentieth century it is
beings is that they are caught in structures with- best known as the theoretical underpinning of
out human intentions or purposes. Where both British social anthropology, as exemplified
people are portrayed as having everything from by the work of Malinowski and Radcliffe-Brown,
their consciousness to their material conditions and of the American structural functionalism of
dominated by the place in the structure that they Parsons and Merton. More recently, functional
were born into, all that is necessary is some no- explanation has enjoyed a renaissance in the
tion of Fate. As Kovel (1976, p. 223) states: work of the structural marxists, Althusser, Castells,
and Poulantzas (Burawoy 1978). The links be-
Man is homo laborans, defined by his situation in tween the functionalism of the structural marxists
the labor process; his psychology is defined then by and that of the structural functionalists are quite
such traits as would make him want to work, pro-
close, although their views are strongly opposed
ductively to transform nature, or to exploit those who
do work. Depending upon the objective situation, on certain issues. In this regard, it is interesting to
one or another side of these propensities will come note the debt that Poulantzas ( 1979) acknowl-
to the fore, giving to Man his place in the class strug-
edges to Parsons. Similarly, Harvey (1978b, p.
gle. To this-which of course corresponds to the 108) writes in discussing a diagram of the circu-
term "vulgar Marxism"-all other attributes are
lation of capital: "The diagram looks very
mere incidentals, residual categories at best, or
positive mystifications. To such an eye, man is a 'structuralist-functionalist' because of the method
passive creature whose consciousness is simply de- of presentation. I can conceive of no other way
termined by the impingement of material forces. to communicate clearly the various dimensions
of capital flow."10
A major difficulty with a diminished or absent Functionalism is a type of holistic explanation.
concept of consciousness is that it leads to mys- Its basic premise is that wholes, whether they be
tification in any discussion requiring reference to living organisms or societies, are composed of
cognitive processes. The following quotation parts that both form the whole and contribute to
serves to illustrate this tendency. "Labor seeks its its maintenance and development. In order to
own meanings, partly derived from a rapidly do this, functionalists try to "grasp the fullness of
fading memory of artisan and peasant life, but the interconnected, living, mutually adjusting
also from the ineluctable imperative to learn nature of a social system" (Skidmore 1979, p.
what it is to be human" (Harvey 1 978a, p. 34). 1 1 6). For the functionalist, both in anthropology
The mysterious forces here are the notion of a and in sociology, the function of any given ac-
"collective memory" that labor as a class is sup- tion is the effect which that action has. Usually,
posed to have and some notion of a "collective functional theory is an attempt to account for
drive" or "need." What is this collective memory regularities in behavior that do not appear to re-
and what are the processes involved in its oper- sult from the conscious desires of individuals (Ryan
ation? What is this need to learn what it is to be 1970). Explanation is framed in terms of the
42 Duncan and Ley
properties of a social whole, whose purposes system needs to produce an industrial reserve
are served by the actions of individuals. Func- army, that the function of welfare is to permit
tionalists emphasize the concept of system, for it unemployment, and that the function of ideol-
captures the interrelatedness of wholes. It pro- ogy is to make people think that they have free
vides a context in which one can trace the func- will (Peet 1975, p. 564; Blaut 1974, p. 40; Short
tional effects of action. Santos (1977, p. 4) rep- 1 976, p. 80). Again, we must stress that this is
resents this view among the structural marxists not functional for individuals per se but for the
when he argues that "society evolves system- needs of the system. The system through the
atically" and "socioeconomic formations can be mechanism of ideology, which as we saw
understood only within the framework of a total- above is functional to it, takes individuals and
izing movement, all of whose elements are simply "inserts them into their practical activities
variables interacting and evolving together, supporting this structure" (Short 1976, pp.
governed by the law of the whole." Castells 80-81).
also makes extensive use of systems terminology Functional theorists have traditionally pro-
and complex flow diagrams in his abstract disposed an equilibrium model whereby a system
cussion of contemporary urbanism. Where does is thought to evolve toward equilibrium. It is ar-
the justification of his abstract system lie? Ironical- gued that the system contains mechanisms that
ly, he turns to the same device as conventional protect its integrity as a system and maintain
systems theorists like Jay Forrester: "The technique stasis. The marxist version of functionalism, on
of experimental verification that seems most the other hand, rejects equilibrium theory at one
adequate is that of a simulation model" (Castells level and suggests that systems are nonequilib-
1977, p. 274). rioting. This disequilibrium in the system is
Entities such as "society," "capitalism," and achieved by the inescapable presence of con-
"the social formation" are viewed as systems, tradictions. It is held that the clashes between the
which are integrated with their own "logic" and contradictions in the functional needs of the
set of functional prerequisites or "conditions of re- ,capitalist system prevent it from achieving equi-
production," whereby they reproduce themselves: librium and thus push the system forward on its
"historic mission"-to a higher level equilibrium
The totality seeks to shape the parts so that each
part functions to preserve the existence and general
(Preteceille 1976, p. 73; Blaut 1975, pp. 1, 6;
structure of the whole. Capitalism, for example, Peet 1975, p. 566; Peet 1977a, p. 257; Harvey
seeks to shape the elements and relationships 1975b, pp. 13, 17; Walker 1978, p. 169).
within itself in such a way that capitalism is repro- Functional explanation has usually been crit-
duced as an ongoing system. Consequently, we
icized for being tautological. Functional theorists
can interpret the relationships within the totality ac-
cording to the way in which they function to pre- assume that a functional model explains the
serve and reproduce it (Harvey 1973, p. 289). nature of a system. In other words, they believe
that things exist in order to fill the needs of a
There is further reference to the minimum re- system. And how does one know that some-
quirements of capitalism in order that it survive thing fills the needs of the system? Because it
the functional working out of the components of exists. A function exists, therefore, because it is
capitalism, and the need of social formations to functional, and we know that it is functional be-
recreate themselves (Piclkvance 1976a, p. 59; cause it persists. This tautology, then, does not
Harvey 1972, p. 10; Harvey 1976, pp. 81, 88; constitute an empirical proposition or an
Harvey 1978a, p. 14; Walker 1978, p. 168; explanation-it is a definition. Armed with the
Peet 1975, p. 564; Gray 1976, p. 39; Wisner notion of functionalism, the theorist's research
1978, p. 90). In addition to these general func- program consists of trying to show how every-
tional relationships within the system, the fol- thing from "the physical landscape," to "in-
lowing phenomena are held to be functional to equality," to "geography" is functional. Tradi-
capitalism or to the social formation: housing, tional functionalists have usually encountered
education, urbanization, the physical land- great difficulty demonstrating that everything is
scape, and geography (Gray 1976, pp. 38, 42; functional, for there are certain elements in a
Walker 1978, pp. 168-69; Harvey 1978a, p. system that appear to be dysfunctional to its
9; Peet 1979, p. 165). We are further told that continuation. Marxists who subscribe to a func-
inequality is functional to the system, that the tional perspective have no such trouble because
those elements which are dysfunctional to the "final stage." "Consequences at one stage be-
stability of the system are defined as "contradic- come preconditions at the next." Or the process
tions," and are seen as being functional to a may be described in botanical terms: "The petty
brooder goal, for they move history to a higher commodity mode of production held within it-
plane. In short, they have gathered in some of self the seeds of its own transformation to indus-
the loose ends of functionalism and rendered trial capitalism, and as it unfolded . . ." (Blaut
the tautology logically tighter. By making dys- 1975, p. 6; Harvey 1976, p. 88; Blaut 1975, p.
functional elements functional simply by 2; Harvey 1975b, p. 14; Walker 1978, pp.
changing their context, they have removed any 173-74). This evolutionary process is typically
possibility of challenging the theory on empirical characterized by conflict: "Capitalism is a system
grounds. Everything is by definition either func- propelled through time by the eruption of its in-
tional to the system or a contradiction that is herent contradictions" and, "we have entered a
functional to the dialectical evolution of the sys- dynamic propelled by the mass force of colliding
tem. Such a theory is perfectly logical but utti- material events" (Peet 1977b; Peet 1977a, p.
mately vacuous for there is no way of knowing 260). These contradictions of dysfunctional ele-
whether or not it is correct. It also raises the issue ments within a system cause the system to go
of teleology, the purpose behind the system. out of equilibrium and thereby move it forward
In places, Marx seems to have believed that down its "historically determined course." After
he had uncovered a logic or telos to history, and the "mechanism of history" has been identified,
this belief remains one of the central tenets of disequilibriating events are simply explained.
marxism (Burawoy 1978, p. 60). History, it is ar- The mass student demonstrations in the United
gued, has a purpose, an end to which it is mov- States during the 1960's, for example, "resulted
ing. Just as there is a mission to history, so given from the maturing state of contradiction into
the functionalist mode of explanation, classes which capitalism was moving," while Piclkvance
have a mission to act out the functional needs of tells us that "state intervention, in Preteceille's
society at any given stage in its historic devel- felicitous phrase, 'is not a response to the con-
opment. The bourgeoisie, we are told, must ac- tradiction as such but a contradictory response to
complish its "historical mission," the accumula- the terms of the contradiction' " (Peet 1977a, p.
tion of capital. The capitalist's participation in this 241; Piclkvance 1976a, pp. 62-63). All of this
"historical mission" does not stem from his own contradiction finally makes capitalism fall into "a
will, but from the need of capitalism (Harvey state of ultimate contradiction" and history
1975b, pp. 9, 20). Clearly then, this historical moves on to its next stage (Peet 1977a, p. 261).
mission has nothing to do with people's con- This teleological explanation implies that
sciously trying to "shape history." The movement there is some Rationality which is guiding events
of these systems through history is conceived of in the world to some preordained goal. This is
as evolutionary change. Capitalist social forma- not the rationality of individuals but of the total-
tions evolve, and the elements of systems ity. We are arguing, of course, that the position
evolve, often at different rates (Harvey 1975b, of these structural marxists is highly idealist.
p. 14; Gray 1976, p. 42; Blaut 1975, pp. 5, 12; Perhaps this can best be illustrated by quoting
Santos 1977, p. 6). Quite clearly we can see the from Hegel (1953, p. 89) on how the purpose-
reliance here upon an organic analogy; social ful movement of history is accounted for by the
organisms can be thought to evolve in the same dialectic of conflicting forces:
manner that natural organisms do. This
evolutionary process, according to the marxists, The Spirit, devouring its worldly envelope, not only
takes place in stages: "All societies will pass passes into another envelope, not only arises re-
juvenated from the ashes to its embodiment, but it
through a definite set of historical stages." At
emerges from them exalted, transfigured, a purer
work here are a set of "historically determined
Spirit. It is true that it acts against itself, devours its
laws" that create an "historically determined so- own existence. But in doing so it elaborates upon
ciety(Blaut 1975, p. 10; Santos 1977, pp. 3, 5). this existence; its embodiment becomes material
Just as "nature" is guided by a set of laws, the for its work to elevate itself to a new embodiment.
laws of evolution, so is society. The organic ter-
minology pervades the discussion of the stages In Hegel's account of the process of history we
of history. Capitalism "matures" and enters its can see concepts similar to those the structural
44 Duncan and Ley
marxists have used: that history has a purpose, cost of a reductionism that sees the world only in
that it evolves from stage to stage and that the economic categories-which is to commit the
seeds of the future stage are contained in the methodological error of economism.
former, that the process is fueled by contradic- It is a telling coincidence that some of the
tions arising within each stage, and that there is leading protagonists of a marxist perspective in
an underlying rationality or functionalism to this geography were equally confident in their es-
whole process that keeps it on course. Hegel's pousal of positivist science in the 1960's, a science
philosophy is profoundly compatible with that of that Harvey (1973, esp. Chap. 4) has correctly
the structural marxists, only the words have associated with the interests of rational, techni-
changed. But perhaps the philosophy of struc- cal, and (in his terms) bourgeois control.11 The
tural marxism is just that, a mere verbal trans- leading sector of positivist application in human
formation of Hegel's claim that the Idea guides geography was spatial analysis, and the under-
the course of history. pinning of spatial analysis has been neoclassical
economics. Neoclassical training has proven
particularly fertile for the development of struc-
tural marxism, for the shift in emphasis is not as
Marxism as Social Science
extreme as might be expected. The reversal by
structuralists of the categories of neoclassical
If Hegel represented an apparent counter- economics does not lead to their transcen-
point to Marx's philosophy, then British political dence, for both share a fundamentally eco-
economy provided the seeming counterpoint to nomic and narrowly rational view of the world.
his social science. Even in his early writings, a When Harvey produced his answer to the ghetto
major objective was "a careful critical study of problem, it was an answer that remained within
political economy" (Marx 1964, p. 63). Main- the conceptual realm of land economics; as one
taining the distinction within the marxian tradi- critic suggested, to make the von Thonen or
tion raised earlier between political economy Alonso model not true, as Harvey wanted, is not
and social history, it is the former genre which to escape their categories and view of the world
usually embodies philosophical holism and its (Harvey 1973, p. 137; Olsson 1972). Elsewhere,
attendant weaknesses concerning a passive there is the acknowledgment that in the United
model of man, the problematic relations be- States the process of the competitive bidding for
tween consciousness and structure, and func- urban land "is, in most respects, identical to that
tionalism and teleology as properties of the assumed in neoclassical models of land-use de-
internal logic of systems. As a result, within marx- termination" (Harvey 1978a). The grounds for
ian political economy one commonly finds the Peet's earlier sympathy for the von Thunen
displacement of the errors of an undergirding model, which has been widely adopted by neo-
holistic philosophy. As it is the political economy classical economists, bears striking resemblance
tradition that dominates marxist social science in to his contemporary structuralist stance. In 1969
geography, this section will chart the transmis- he promoted the von Thunen model, as "it al-
lows a logically connected system of forces to be
sion of structural affirmations into empirical work
and indicate the unresolved quandary that has advanced as the economic structure underlying
resulted. the events portrayed" (Peet 1969). The allu-
sions to the logic of system, impersonal forces,
and deep economic structures lying beneath
surface manifestations are equally central to his
The Reproduction of Economic Categories
apparently reformulated structuralist position.
It is not difficult to duplicate examples that
The embeddedness of marxist geography in show how conventional and marxian econ-
political economy has two consequences we omists often share the same conceptual do-
shall now elaborate: firstly, the break with con- main, even if their interpretations are reversed.
ventional social science is less complete than is The treatment by radicals of business cycles
commonly asserted, and secondly, the claim to and urban development (Walker 1978) re-
a universal set of insights across "the totality of vives much earlier studies by such bastions of
society" is invalid, and may be made only at the establishment economics as Homer Hoyt (1933)
and Paul Samuelson (1942). In rejecting much categories of political economy is to make over
of contemporary urban sociology, Castells the whole world into a giant workhouse.
(1977, p. 123) finds himself most sympathetic It is particularly in the realms of cultural rela-
to human ecology as occupying a conceptual tions and everyday experience that one finds
realm not incompatible with his own. But "the real silences of Marx," and these silences
human ecology is surely the least social branch are shared both by land economists and their
of sociology. Park's alternative term for the sub- marxist critics (Thompson 1978b, p. 357). In
field was biological economics, a label that in- Alonso's (1 965, pp. 1, 17) neoclassical land-use
timates both its essentially Darwinian focus upon theory, personality enters the scene only as "a
impersonal, biotic forces and its allegiance to pallid skeleton. . . reduced to that uninteresting
neoclassical economics. Moreover there is a con- individual, economic man." More complex
sistent root to human ecology: "To a large ex- sociocultural models of man and the perceived
tent the model may be seen as one of eco- environment "are undoubtedly important, but
nomic determinism" (Timms 1971). As we will no way has been found to incorporate them into
see, the issue of economism continues to be an the type of theory that will be presented"
abiding source of confusion in marxian theory (Alonso 1965, p. 17). So too, the human
also. ecologists rejected any consideration of cultural
The overdependence on economic cate- values in their discussion of land-use patterns,
gories, in his case derived in large part from though "not because they are unimportant, but
British political economy, has been criticized in because the assumptions and point of view of
Marx's own work. The French theorist Baudrillard human ecology are not adapted to their treat-
has attacked historical materialism as "too con- ment" (Quinn 1940). Implicit in these state-
servative, too rooted in the assumptions of po- ments is an anti-humanism that is based on an
litical economy, too dependent on the system internal epistemological shortcoming, as the
of ideas it seeks to overthrow . .. the difference logic of mechanism is conceptually unable to
between Marx and political economy is not as accommodate the social psychology of lived
great as their agreement" (Poster 1 975b, pp. 2, experience. In structural marxism these epis-
5). So too, Thompson (1978b, pp. 251 -52) temological limitations of economic categories
sees in Marx's encounters with political economy are either justified or rationalized theoretically.
something that is "obsessive. .. what we have For example, the theoretical importance of the
at the end, is not the overthrow of 'Political moral order of lived experience is repressed in
Economy' but another 'Political Economy.'" favor of the logic of economic structure; percep-
The dominantly economic conceptualization tions and preferences are discussed as "purely
of the world becomes problematic when it is epiphenomenal"; the notion of urban subcul-
overextended and treated as more than a par- tures that exercise real effects is rejected as "a
tial perspective, for "a capitalist mode of pro- myth rather than a specific social process" (Har-
duction is not capitalism" (Thompson 1 978b, p. vey 1973, p. 131; Roweis and Scott 1978;
346). To collapse the range of social experience Castells 1977, p. 95). Even institutions are re-
to the outworking of deep economic structures is duced to impotence: "Individual institutions are
to present an impoverished view of the social, simply agents of the underlying economic pro-
cultural, and political realms of life. Thus cess. They may modify particular aspects of that
capitalism is regarded not as an economic sys- process but as agents they cannot change it or
tem but as broadly "as a mode of existence"; be held responsible for it" (Lee 1979). Before
urbanism is diminished to "a product of the cir- the transcendental object, people assume an
culation of surplus value"; while the city is iden- essentially passive form, and cultural relations, if
tified as merely "a residual unit of labour power" they appear at all, are granted a predictably
(Peet 1978; Harvey 1973, p. 312; Castells derivative status, as they dwindle beneath "the
1976). Social problems in the city are treated as phantom of production" (Baudrillard 1975, p.
"simply the surface manifestations of a structural 17). It is not apparent how such views differ sub-
process of change which can be understood only stantially from Althusserian structuralism where,
at the level of the mode of production" (Lee as we saw earlier, men and women become
1979). One is reminded of Adorno's complaint passive carriers of a structural logic they cannot
that to conceptualize problems only in the resist: "The true subjects. . . are not . . I'concrete
46 Duncan and Ley
individuals'-'real men'-but ... the relations of then abolishes geography. If capital can create
production (and political and ideological social landscapes at will, why does it systematically
relations)" (Althusser and Balibar 1970, p. 180). select certain sites rather than others? Why are
The theoretical requirement for the reduction only certain neighborhoods (with a central loca-
of human choice and culture identifies structural tion and high levels of architectural and environ-
marxism as a mirror image of the equally ex- mental amenity) and certain cities (major cities
treme neoclassical concept of consumer with growing downtown quaternary employ-
sovereignty. Having dismissed the role of con- ment) chosen? The answer is that capital cannot
sumer choice in the creation of urban land val- create landscapes at will; its deployment is
ues, Roweis and Scott (1978) turn to "the system heavily constrained by local geographies and
of production," the "deep structure" of the spatially biased patterns of existing and
capitalism. So too, inner-city gentrification is re- anticipated demand.
garded as being independent of consumer A necessary corollary of an impoverished view
demand; rather it functionally follows "the of market demand, cultural relations, lived ex-
needs of capitalism." From this perspective, perience, and human volition is an overly de-
'gentrification is a back to the city movement all termined view of economic categories. As some
right, but of capital rather than people." It is writers favorably disposed to marxism acknowl-
capital not consumers which generates demand edge, economism has been and continues to
surfaces; people are merely puppets "taking be an abiding defect in marxian analysis.
advantage of this returning capital" (Smith Thompson (1 978b, pp. 359-60) rejects a per-
1 979b). A theoretical reversal in the positions of sistent economism which has reduced history to
supply and demand has occurred; marxist analysis "a conceptual meccano set"; Hirst (1 977) notes
has its affinities with supply-side economics. how economism has been "the dominant basic
These structuralist arguments are misleading form of Marxist theory"; Poster (1 975a, p. 43)
for two principal reasons. Firstly, the rejection of comments how "Marxism had a tendency to
individual and social demand factors and the become distorted in this specific direction, a ten-
elevation of the reified power of capital do not dency that has yet to be explained ade-
square with the geographical evidence. "Capi- quately"; Johnson (1978) observes that such
tal" is for more conservative than the structuralists
models of mechanism "are astonishingly persis-
imagine, and inner-city reinvestment, for tent, not least within Marxism"; while Clarke
example, has been perceived as too risky for (1979) sees marxism as "still dominated
entrepreneurs until the power of market de- theoretically and politically by an economism
mand has established itself. Indeed the record that seeks to deny the creative and imaginative
shows that rather than expediting gentrification, powers of the working class." Before the weight
institutional gatekeepers have raised persistent of such sympathetic critics it is not enough to
obstacles during the early critical phases: finan- allude to a few of Engels's final letters to divert
cial redlining, developer scepticism, and dys- the charge of economism, as if these fifteen
functional by-laws have all been impediments pages of private correspondence carried some
that nonetheless have not thwarted would-be superior authority to the published life work of
gentrifiers. In some cases of revitalization (nota- him and Marx (Gouldner 1980, esp. Chaps. 8,
bly Society Hill in Philadelphia) the state has 10). A more accurate representation of the im-
been an active source of investment from the passe is offered by Stedman Jones (1979), a
beginning, but this need not imply the theoreti- marxist historian influenced by structuralist ex-
cal primacy of "capital," unless the state is itself planation:
treated as simply an expression of "capital."
Moreover, even this contingency (were it true)
Economism is not some arbitrary deviation from
would not require a structural explanation, for it Marxism, introduced by Kautsky, Stalin, or whoever
could be accommodated equally within con- else on whom one may wish "to pin the convict's
ventional elite theory. There is a second weak- badge." It is there from the very start, from the
German Ideology onwards. From 1848, Marx and
ness in the structuralist argument. To propose
Engels were aware of the problem, like so many
that "capital" operates as a formal cause in
Marxists since. But it is a vain endeavour to imagine
gentrification is to ignore the realities of areal that they overcame the problem, that the solution is
differentiation; like the isotropic plain, capital there, if only you read the text in the correct way.
They did not resolve the problem, any more than to the whole Marxian theory of society" (Homil-
we have.
ton 1974, p. 22). But it is a term that is well
Geographers and urbanists in this tradition defined only at the level of theory in terms of
have been no more successful in transcending functional positions relative to the mode of pro-
economism. For not only are there such un- duction. The theoretical analysis in Capital rep-
equivocal assertions as "the material economic resents, as Marx acknowleged, a "pure form" of
base of society determines the superstructure of capitalism. Like the physicist, he is looking at
social, legal and political institutions" or, "when phenomena "in their most typical form and
we attempt to view society as a totality, then most free from disturbing influence" (Giddens
ultimately everything has to be related to the 1973, p. 33). But if the theory represents an
structures in the economic basis of society" "ideal average" of capitalism, what then is the
(Piclkvance 1976b, p. 1; Harvey 1973, p. ontological status of such concepts as social class
292). 12 More pervasively, there is the attempt to (Tristram 1975)? Hall considers that Marx's des-
cast explanation continually and everywhere in ignated classes are "personifications" without
terms of economic imperatives, leading as we a consistent empirical equivalent; their status is
shall see to a crisis in empirical exposition. functional rather than empirical (Hall 1977). But
The dependence on economic categories in then to identify whole classes empirically from a
analysis has an ironic outcome. "By presuppos- functionalist derivation about the mode of pro-
ing the axiom of the economic, the Marxist duction becomes "an untenable proposition"
critique perhaps deciphers the functioning of (Hall 1977). Marx's own historical studies indi-
the system of political economy; but at the cate the problematic ontology of his category.
same time it reproduces it as a model" (Baud- In Louis Bonaparte's France, instead of the
rillard 1975, p. 66). Structuralists in geography dichotomous or tripartite class structure required
have reproduced many of the categories and by theory, half a dozen class interests were iden-
the shortcomings of spatial analysis; their tified (Marx 1954, pp. 17, 28). Even in England,
rationalism, formalism, abstraction, and partic- which Marx considered the "classic ground" of
ularly their preoccupation with economic capitalism, classes were not ranked as theory
categories are reminiscent of the workings of the required (Giddens 1973, p. 33). In the un-
bourgeois mind. 13 finished fragment offering a definition of class at
the end of the third volume of Capital, there is
the observation that in England "the stratification
The Problem of Marxian Categories:
of classes does not appear in its pure form. Mid-
Social Class
dle and intermediate strata even here obliter-
The preeminence of a priori theoretical ate lines of demarcation everywhere (although
categories is a feature of structuralist method; incomparably less in rural districts than in the
moreover, these categories are not treated as cities)" (Marx 1967a, Vol. 3, p. 885).
arbitrary but as necessary, demanded by the What is remarkable about this fragment is that
logic of explanation and therefore, one as- it was precisely in the British industrial cities that
sumes, implicit (if not explicit) in actual cir- Engels claimed to identify so unequivocally the
cumstances. But in the theoretical distillation polarized presence of the two classes, which he
away from empirical events, concepts undergo read off in Manchester from the map of residen-
stasis, an ossification that may leave them tial segregation in a passage much cited by
stunted with the passing of the circumstances for geographers. "In these towns," he noted, "there
which they may have been relevant. As a result are only rich and poor, because the lower mid-
the fit often seems awkward or forced when dle classes are fast disappearing. At one time
they are applied in an analysis of actual this section of the middle classes was the most
societies. The interlocking is by no means as stable social group, but now it has become the
satisfactory as theory would suggest. Con- least stable" (Engels 1958, pp. 28-29). But it
sequently, categories are either imposed un- seems as if, despite the cautions of his Preface
compromisingly onto events, or else their defini- against uncritical theoretical foreclosure, Engels's
tions become imprecise, and it is through this own theory had outpaced his fieldwork. David
Ward (1976; 1980), for example, has chal-
purely linguistic device that the argument is carried.
Class is a concept whose definition "is central lenged the dichotomous scheme of class segre-
48 Duncan and Ley
gation that Engels perceived in Manchester, p. 101). The new middle class has been as-
noting that although there were concentrations signed by different authors alternatively to cap-
of the elite and the destitute in the British indus- ital and to labor, culminating in the suggestion
trial city, "far larger areas housed a mixture of that functionally they belong to both (Parkin
lesser professionals, petty proprietors, master 1979). Class boundaries have therefore been
craftsmen, journeymen, laborers, and domestic drawn above the middle class, below it, and
outworkers. " through it, a highly unsatisfactory treatment of
More significant than the residential map, the largest single grouping in the advanced
however, was the social structure, and here economies. Moreover, and this can only be
there is little evidence of the exclusive class divi- mentioned here in passing, there are other lines
sions Engels was looking for. Forty-five years of stratification, following ethnicity, sex, religion,
after his research in Manchester he lamented and divisions according to consumption styles,
that rather than class divisions becoming more which frequently act as the basis of perceived
polarized as he had predicted, they had be- interests and group action (Parkin 1979; Saun-
come far more confused (Ward 1975). The ders 1978).
working class was characterized by innumerable In light of the major limitations to the exten-
gradations and subcultures. The lower middle sion of a simple theoretical model of class to
class, rather than fading away, has made a empirical conditions, it is disappointing to note
powerful resurgence with the mushrooming of precisely this literalism in many geographical
government and the tertiary sector. 14 The treat- studies. Harvey (1 978b) extends to capitalist so-
ment of the middle class by Engels is equally ciety the dichotomous class system found under
unsatisfactory. Neither shopkeepers nor profes- the pure form of capitalism: "The class character
sional groups appear in Engels's Manchester.15 of capitalist society means the domination of
His monolithic grouping of the middle class is labour by capital." Elsewhere we read that
unconvincing and, even to a sympathetic critic, "conflicts in the living place are, we can con-
the vilification of them is too complete and un- clude, mere reflections of the underlying tension
qualified (Marcus 1974, esp. pp. 238-39). The between capital and labor" (Harvey 1978a;
irrationality of this conceptualization of the mid- also Cox 1981). Harloe (1977) is on firm ground
dle class is clarified in one illuminating footnote: when in a review essay he identifies the
"I have not had time to differentiate between dichotomous class model derived from produc-
the different sections of the bourgeoisie or the tion relationships to be "the fundamental re-
various middle-class political parties. This subdivi- lationship" in a radical political economy of
sion of the middle class is only of historical and space. As we will see, such literalism in the
theoretical significance" (Engels 1958, p. adoption of categories leads to major confusion
331, emphasis added). But if both history and in empirical research.
theory are inconsequential then what remains? There is a second solution to the problem of
A political ideology? But if the whole concep- categories. It has often been noted that there is
tualization is ideologically determined, then any considerable looseness and inconsistency in
talk of science is a charade. Marx's own use of class, as his theoretical work
The fracturing of any dichotomous class system impinged upon historical events. 16 Rather than a
about the mode of production in the advanced literal extrapolation from theory to the evi-
societies has become more nearly complete in dence, a looser definition permits a more diffuse
the twentieth century. The dissolution of owner- identification of social groups with classes, as the
ship in joint-stock companies, the awesome criteria for class membership are varied accord-
power of life insurance and pension funds, the ing to the context. But this raises two further dif-
growth of public-sector and nonprofit occupa- ficulties. The first is terminological vagueness.
tions in the economy, the increase in automa- Consider the definition of social classes as "com-
tion and technology, the expansion of the binations of the contradictory places defined in
white-collar sector, and the decreasing role of the ensemble of the instances of the social
unskilled labor, are some of the historical trends structure" (Castells 1977, p. 243). This is a phrase
that are intractable problems for a dichotomous so imprecise that it will obligingly bend in any
class model. In particular, the emergence of a direction the theorist may choose. A second
new middle class "has always escaped ade- issue is also at stake. To follow the evidence
quate analysis in Marxist terms" (Giddens 1973, more closely and develop a more plural and
elastic concept of class may not require refer- licized over the past twenty years; now it is nec-
ence to the theoretical structure at all. This ten- essary to emphasize also the shortcomings of
dency is marked in Massey and Catalano's highly theoretical inclinations, where the
(1978) study of land ownership in Britain, where theoretician "tends to impose his theoretical
a theoretical discussion of rent and class "frac- categories on historical reality, that is, he loses his
tions" from a structural perspective appears un- openness to the historical phenomenon-in-itself,
necessary to the ensuing empirical material on and he uses the historical reality to illustrate his
land ownership and conflict between plural elite theory, thus reversing the desirable relationship
groups. The same observation holds in Castells's of these terms" (Spencer 1977). These errors,
more recent work, where he repudiates his ear- which we shall document below, give rise to two
lier "codification of history according to Marxist further problems. Firstly, the shortcomings be-
schemata" (i.e., conceptual stasis). Instead, he come critical whenever theory makes the claim
favors "studies where the recognition of a new of practical or political efficacy. Secondly, they
historical perspective is more important than a present an idealism, where the ideas, the a
formalist theoretical orientation" (Castells 1978, priori categories, of the researcher are granted a
p. 12). In these terms, class structure is now de- privileged status over empirical circumstances
fined by income and status categories, and the (Ley 1978). In structural marxism the idea of
ability to gain access to key private and public capital has succeeded Hegel's world spirit; "cap-
gatekeepers (Castells 1978, pp. 21-24). But ital has become Idea, which unfolds itself in his-
this is to replicate Weber's stratification trilogy of tory" (Thompson 1978b, p. 253).
class, status, and power, rather than to require The tension between theoretical formulations
reference to deep structures or a set of produc- and geographical circumstances is a common
tion relations. The suspension of a literalism in feature in the structuralist analysis of advanced
following the theoretical category of class is re- societies. So too is conceptual stasis, the imposi-
placed by a heterodoxy in which structural tion of predetermined categories upon empiri-
theory is either largely redundant or else a con- cal circumstances, and also the instrumental use
trived addition to empirical exposition. of facts as illustrative devices rather than as a
These weaknesses in analysis using the marxist means for formally testing or confirming theory.
category of class are an instance of a more gen- Harvey, in particular, is aware of the difficulties in
eral problem, the unresolved tension between forging the link between theory and history. In
structural theory and empirical investigation. his discussion of Marx's theory of capitalist ac-
cumulation, he notes that it applies "in a 'pure'
state without reference to any particular histori-
The Tension between Theory cal situation." To extend the theory empirically is
and Empirical Study then not easy: "We have to force an intersection
between the theoretical abstractions, on the
The friction between abstract theorizing and one hand, and the materialist investigations of
empirical study has recently come to the fore in actual historical configurations on the other."
a series of interchanges in British social history.17 Moreover in this task Marx himself provides little
In part this has been a quarrel over spheres of precedent, for "we have to accomplish also that
influence on the margins of history and sociol- transformation from the general to the concrete
ogy; but, more significantly, debate has re- which comprised the central thrust of Marx's un-
volved around the relative status of theoretical finished work" (Harvey 1975b). Notable in
and empirical research in historical explanation. these extracts are the methodological sequence
Sociologists have asserted methodological from the general to the particular, the substan-
sophistication, rigor, and theoretical refinement; tial problems involved in forging this union, and
historians have claimed clearer contextual un- the implication that to date these connections
derstanding, more respect for the facts, and less have not been made successfully. We shall now
assurance of the appropriateness of elaborate examine in more detail the full measure of
theory. Charges of abstraction are met by these difficulties, noting how the literal attempt
countercharges of empiricism; the tension is that to transpose theoretical categories onto geo-
"of the universal versus the concrete" (Spencer graphic circumstances is awkward and rarely
1977). In geography the limitations of idiosyn- convincing, while too close an attention to the
cratic local studies have been effectively pub- facts invites the charge of heterodoxy in its de-
50 Duncan and Ley
parture from the theoretical mold. This is one missed with the assertion that the suburbs can-
aspect of the contemporary "crisis of Marxism," nota meet "the real human requirements of la-
crisis that is largely concealed in the optimistic bor" (Harvey 1978a).18 What are these human
pronouncements of marxist geographers needs? How may they be recognized? By which
(Gouldner 1980, esp. pp. 26-29). processes may they be satisfied? And who is
"labor" anyway?
The explanation of land-use conflict is no
Empirical Confusion: Literalism more convincing (Harvey 1978a). Again the
exposition begins, "for purposes of analytic con-
The shortcomings in the substantive studies by venience," by assuming the clear division of
structuralists may be considered in more detail in labor and capital, rather than demonstrating it
the work of Harvey and Castells, perhaps the empirically. This theoretical primacy asserts what
two most influential marxian commentators on should be demonstrated: "If we derive our con-
the city in advanced societies. Harvey's more re- cept from a prior theoretical model of a struc-
cent research and Castells's earlier work (now tural totality . . . we will suppose that class is
largely repudiated by Castells himself) usefully instantaneously present (derivative, like a geo-
indicate the difficulties of a literalist position (Ley metric projection, from productive relations) and
1979). that hence classes struggle" (Thompson 1 978a).
Harvey's discussion of residential segregation The status of historical reality then becomes
begins with his familiar acknowledgment to merely illustrative, as Spencer (1977) suggested
Marx's system and a symptomatic declaration of is normal in such cases of theoretical overelab-
the fundamental nature of a dichotomous class oration. In the text, real events are introduced
division. Within this system, residential differ- only as illustrative allusions to bolster the
entiation has "an integral mediating influence in theoretical framework. However, although the
the process whereby class relationships and so- stated objective is to illuminate conflicts under
cial differentiations are produced and sustained" contemporary advanced capitalism, the
(Harvey 1975a). The argument is highly theoretical argument rarely intersects with pres-
functionalist, as it proceeds by positing that ent geographical conditions. Instead, specific il-
segregated working-class neighborhoods serve lustrations (which are few in number) are biased
to perpetuate occupational skills for the needs of toward early rather than advanced capitalism,
future generations of employers, and that, as drawn from "early industrialists": the introduction
these spatially constrained households establish of Henry Ford's eight-hour day in 1914, the
community, so they forge common tastes and Pullman strike of 1894, the plight of the Lowell
preferences for the convenience of producers, mill girls in the 1.820's. In addition, members of
while also limiting aspirations for upward mobil- the worlkforce are repeatedly represented as
ity by shaping a class norm. Thus capital appears factory employees, though all blue collar work-
as a single clairvoyant (the omniscient world ers now account for only 30 to 35 percent of the
spirit?), which simultaneously provides itself with labor force in North America. Indeed, by using
a present and future labor force, a pliant group the strict definition of class set out by one influen-
of consumers, and a subordinate and passive tial structural marxist (Poulantzas), fewer than 20
underclass, all aided by its manipulation of spa- percent of the present-day American labor force
tial pattern. Capital, it seems, cannot err, for if would qualify as members of the working class,
residential dispersion rather than segregation is while over 70 percent would consist of the new
encouraged, by opening up the suburbs to dis- petty bourgeoisie (Wright 1976). The theoretical
advantaged groups, then this too is functional to prerequisites of structuralism have drawn Har-
its interests for it will serve to dissipate latent class vey's argument away from advanced capitalism
consciousness (Harvey 1978b). Moreover, the and toward a time and a place where it may
suburbs themselves represent no more than "a apparently be sustained more easily. In contrast,
Faustian bargain" to labor, an assessment at less theoretically determined analyses of con-
variance with objective and subjective indicators temporary land-use conflict emphasize the
that reveal a high level of residential satisfaction criss-crossing alliances representing the complex
among suburban dwellers (e.g., Gans 1967; form of modern social stratification (Elliott and
Zehner 1977). But the logic of system requires McCrone 1979). Agitation may cut across con-
that their satisfaction be illusory, and it is dis- ventional class lines as in the case of urban re-
newal, where blue-collar unions often support extreme scepticism toward empirical work. It is
the proposals of financiers and the development "a philosophy of the concept," where "theory is
industry, to be opposed by minorities and liberal about a theoretical object, and this latter is al-
members of the middle class, with the different ready at several removes from empirical reality"
levels of government fluctuating from one side (Poster 1975a, p. 345; Glucksmann 1974, p.
to another according to local and national cir- 115). Castells shares Althusser's disdain for em-
cumstances. These novel alignments point to a piricism, preferring the development of
separate articulation of interests than that predi- "theoretical tools." In a separate paper he has
cated on theories of production; the new al- outlined his position more clearly. "Empiricism is
legiances commonly spring from distinctive atti- not only an epistemological weapon, but also
tudes toward consumption (Ley and Mercer an ideological weapon . . ."; theory must
1980). therefore have primary status, as "a means of
In a third paper that attempts to link capital production of scientific facts" (Castells and Ipola
accumulation with urban growth cycles, the 1976). What then is to become of the status of
same tendency is found toward conceptual stasis the historical situation in The Urban Question?
and an overly determining theoretical structure The response is disquieting: "I prefer to take the
(Harvey 1 978b). There is not space to treat the risk of a certain empirical margin of error in
substance of the argument here, but we may order to clarify the ideas in the perspective I
note the awkward meshing of theory and history have outlined" (Castells 1977, p. 200, em-
that runs through it and the persistent tendencies phasis added). So to what does the argument
to instrumentalism. The paper begins by outlin- become accountable? If not to historical cir-
ing Marx's laws of accumulation-set out in Capi- cumstances, then, it seems, to the internal logic
tal, a theoretical thesis of a "pure form" of of "the ideas in the perspective." It is the
capitalism. The "acid test" of such a theoretical theoretical idea, the logic, that must be de-
formulation, it is acknowledged, is the successful fended against historical facts, so that despite
explanation of real events, but such verification the aspirations to theoretical purity in the struc-
is suspended in favor of the use of illustrative tural system, the ontological primacy granted to
sketches that are "not incompatible with the theory and the concept inevitably draws the ar-
broad outlines of the theory we have sketched gument toward idealism.
in." A genuine testing of the theory is deferred as The result in The Urban Question is an un-
"an extraordinarily difficult task for beyond the happy disjuncture between theoretical abstrac-
scope of a short paper." In conclusion the author tion and geographical circumstances. For
comments on the profound limitations in his example, there is the claim that the emergence
empirical analysis: of the metropolitan region "is the problematic of
the organization of space in the advanced
I have been forced to blur distinctions, make enor-
mous assumptions, cut corners, jump from the capitalist societies" (Castells 1977, p. 29). But
theoretical to the historical in seemingly arbitrary two pages earlier, a demographic table has
fashion, and commit all manner of sins which will shown no important difference in metropolitan
doubtless arouse ire and reproach as well as a good growth between capitalist and noncopitalist na-
deal of opportunity for misunderstanding.
tions. Indeed of the twenty-five world cities in-
But as we have argued repeatedly, the source cluded in the table, the metropolitan area with
of these severe weaknesses lies not in the size of the highest percentage growth over the most
the manuscript but in its literalism in extrapolat- recent decade listed is Peking! Conclusions are
ing a pure structural theory to empirical cir- sometimes no more than assertions at odds with
cumstances. The privileged status of preexisting the evidence. In the case of urban renewal in
theoretical categories imposes a high degree of Paris there is the assertion that renewal areas are
closure upon the argument, and places an im- primarily those where housing has deteriorated
possible constraint on the historical situation. and immigrant and unskilled workers are pres-
Manuel Castells (1977) was not able to re- ent while higher income groups are absent
solve this conflict in his text of nearly 500 pages (Castells 1977, p. 308). But the table on the
in The Urban Question. For Castells the status of some page from which these conclusions are
empirical study is even more problematic. His reached roundly contradicts them. For while
major theoretical source is Althusser's reading of housing quality is associated with renewal, social
Marx, a structuralism that is characterized by an composition takes an inverse form to that
52 Duncan and Ley
specified, with a negative association against research on urban managers, which identified
foreigners and unskilled workers and a positive the some relationships, and which grew out of a
association against upper management and Weberian perspective. The notion of state offi-
professional workers and Algerians! If this is not cials as managers again favored Weber in sug-
bad enough, a correlation is then performed on gesting a growing distinction between political
the size of renewal projects against a list of so- allocation and economic forces. Worse still from
cioeconomic variables. On the basis of these a structuralist perspective, empirical analysis of
correlations hypotheses are stated and generali- public and private gatekeepers showed they
zations made; but unmentioned is the fact that could be and often were sensitized to client or
none of the eleven correlations is statistically sig- demand pressures. Rather than consistently act-
nificant. Such empirical nonchalance contrasts ing in the monopoly manner of a ruling class,
strongly with the seeming intricacy of the their actions could also be reactive to shifts in
theoretical structure. popular opinion and unanticipated market
The structuralists have committed the very er- trends. "Tastes and preferences" were theoreti-
rors that Harvey in his preradical writing correctly cally efficacious after all.
perceived as inherent liabilities in functionalist Consequently, to structuralists the institutional
holism: "The danger that we must here avoid at studies were theoretically tainted. Roweis and
all costs is that mortal inferential sin of erecting Scott (1978) refer Harvey's Baltimore research to
an a priori functional model into full theory with- a manipulated city perspective, i.e., a Webe-
out knowing it and without the necessary con- rian position that emphasizes the role of elites,
firmatory evidence" (Harvey 1969, p. 446). interests, and power, with recourse to constraint
rather than to structural imperatives. Harloe, in
Empirical Confusion: Heterodoxy his favorable review of structural marxism, simi-
larly cautions Harvey and Castells for their em-
The second empirical conundrum, the issue of phasis on consumption, which distances them
heterodoxy, concerns marxian study that shows from the theoretically central realm of produc-
more sensitivity to the empirical situation. The tion and of productive rather than finance capi-
problem here is that the course of explanation tal. This tendency "bears comparison, in some
may require little or no reference to the logic of respects, with previous urban sociology and with t
the theoretical structure. The research then Weberian urban sociology" (Harloe 1977, p. 21 ).
squares equally well with alternative theoretical Attempts have been made to restore or-
traditions (often of a broadly Weberian form), a thodoxy to the theoretical deviations of the in-
circumstance that brings charges of heterodoxy stitutional research. Observing of Harvey's hous-
from structural purists. ing research that "its theoretical insights remain
An appropriate starting point is Harvey's semi- obscure" (i.e., impure), Smith (1979a) has in-
nal study of the Baltimore housing market, verted the roles of production and consumption
which was one of the first empirical studies in in his own work on gentrification in a valiant at-
geography to claim a radical perspective (Har- tempt to reestablish the primacy of production
vey 1-975c). It showed effectively how spatially relationships. Other critics have challenged the
discriminatory practices by various institutions in concept of managerialism more directly,
the financial and property sectors and in gov- claiming that the independent-action of man-
ernment had real effects on the residential areas agers has been exaggerated.20 But what does
of the city. This precedent encouraged a range this imply? Surely there is no suggestion that the
of associated studies examining the role of key managers exercise no control, for the fact of
institutions as gatekeepers in urban devel- control has been amply demonstrated in the lit-
opment. 19 But these studies opened the door to erature. In an increasingly complex and man-
theoretical heresy. The discovery of institutional aged society, the role of public and private
subcultures that exercise real effects make links gatekeepers in the allocation of scarce resources
with Weber's theses on bureaucracy. The recog- cannot be ignored (Pahl 1979). But if the crit-
nition of key institutions involved in the allocation icism suggests, as it must, only that there are
of housing and other services also suggested constraints upon the actions of managers, this
fruitful connections with an emphasis on con- does not discredit a Weberian approach, which
sumption and the liberal theory of elites. These does not advocate unconstrained voluntarism.
linkages were made explicit by Pahl's (1976) Weber's explanatory schema incorporated
criteria of "causal adequacy" as well as "mean- plicating a contingent and dynamic unfolding of
ing adequacy": "Structure sets the conditions history where few outcomes are certain. Both in
and limits, but within that wide field ideas gain its context and in its contingency, what is so con-
an affinity with the interests of a social group, spicuously missing is the logic of a theoretical
and may thus serve to alter the structural condi- system and the imperatives of a mode of pro-
tions, so as to lead to a new framework of action duction. These theoretical bases do not appear
and meaning" (Hamilton 1974, p. 89). as necessary (nor are they systematically intro-
Thus the demonstration of constraints upon duced) in the empirical exposition.
managers (or mediating institutions) need not The some observation has been made of
Marx's
carry the discussion to a structuralist framework. A own historiographic account of The
British study of housing markets in inner Birming-
Eighteenth Brumaire of Louis Bonaparte,
ham reaches the Weberian conclusion that whose style resembles Harvey's essay. Gouldner
"housing opportunities, whether for mobility or (1980, pp. 299, 304) refers to this work as an
improvement, were bureaucratically defined" anomaly, indeed "the classical site of the
(Lambert, Paris, and Blackaby 1978, p. 148). anomaly problem in Marxism . . . [where] ...
The authors then go on to show that the options Reality has dealt theory a dizzying blow." In
open to local managers were themselves lim- short, a separation (not a synthesis) between
ited by higher-order political and economic con- theoretical literalism and empirical heterodoxy
straints. At this point there is an ideological rup- is a feature of Marx's own work. It remains only
ture in the argument, as the authors adopt from to be noted that social historians who draw their
Castells a theoretical ruling-class model of the cues from historiographic work like The
state (a model that Castells (1978, p. 181 ) has Eighteenth Brumaire are constantly challenged
since rejected as an expression of "mechanistic by theorists as to the authenticity of their marxist
Leninism"). But only with this implausible break credentials. Stedman Jones (1979) chides the
from the empirical analysis have the imperatives work of many British social historians who claim a
of orthodoxy in structural marxism been marxian identity: "To call this Marxism demands
achieved. a real effort of imagination." Thompson himself,
At first sight it may seem incongruous that the the sharp critic of the structuralists, has frequently
work of a single author (such as Harvey) might been imputed with such heterodoxy (Johnson
be subject to the criticisms both of literalism and 1978; Hirst 1979; Nield 1980).
of heterodoxy. But the distinction that is drawn This criticism has recently been repeated from
parallels the distinction between his theoretical within geography (Gregory 1981). Gregory
and empirical writing, for if the former may be finds Thompson's schema unduly removed from
described as theoretically overloaded, the latter structural effects, while at the some time he is
makes few essential links with the theoret- properly cautious of the weaknesses of structural
ical structure. This is most clear in the un- theoreticism. This leaves his argument on un-
characteristic essay on "Myth and Monument," certain ground, and he turns in a general way to
a description of the Paris Commune and the precedent of Vidal de la Blache and Gid-
an interpretation of the subsequent construction dens's conceptual framework of structuration
of the Basilica of Sacr&-Coeur as an ideological (Giddens 1979). The French School shared a
statement of hegemony by conservative groups neo-Kantian tradition with Weber, while the
(Harvey 1979). The essay is historiographic, a structuration thesis as presented by Gregory in-
narrative that the author repeatedly refers to as corporates a language of actors, interaction,
"our story." What is notable about the account is routines, interpretive schemes, norms, symbol
its emphasis on context: the fortunes of the mili- systems, domination, and legitimation-which is
tary campaign with Prussia, the power of the to rehearse a thoroughly Weberian vocabulary.
state, the diversity of social classes and interest There is, in short, often a strongly Weberian
groups who flit across the historical scene, and component to apparently marxist analyses that
the range of motivations they express (piety, treat empirical studies more seriously than as
fear, patriotism, etc.). The accuracy of the in- ammunition for illustrating theory. Parkin's
terpretation as ideology critique must be judged (1979, p. 75) provocative comment that "in-
by historians. A separate issue is at stake here. side every neo-Marxist there seems to be a
Beneath the general context there is the imma- Weberian struggling to get out" is worthy of the
nence of particular events and personalities im- most careful examination.
54 Duncan and Ley
For present purposes it is unnecessary to press Conclusion
the convergence with broadly Weberian
streams of analysis; what is of central importance In this essay we have suggested that the inter-
is how dispensable structural theory around the section of human geography with structural
mode of production has become, and how with marxism has led to a passive model of man that
this a distinctively marxist identity to empirical is conservative and results in an obfuscation of
study has been severely challenged. the processes by which human beings can and
do change the world. Furthermore, philosophi-
cal holism is extremely difficult to apply in em-
Validation and the Ideological Imperative pirical research, the result being that in some
cases the explanations are totally inadequate,
Like the positivist tradition that preceded it in with causal power attributed to abstract mental
geography, structural marxism is eloquent in its constructions, while in other cases theoretical
self-proclamation as objective science, "a truly structures are almost completely divorced from
scientific social science" (Peet 1 977a). But this the empirical analysis. In the former instance,
is a suspiciously repetitive claim, which the ontological fallacy leads, as we have shown,
Kolakowski (1978, Vol. 3, p. 486) observes is to the neglect of many types of variables, espe-
not a habit of real scientists. We have com- cially social psychological factors. The Mode of
mented frequently on the discordance between Production or the Logic of Capitalism are mystifi-
theoretical orthodoxy and empirical cir- cations replacing a detailed examination of the
cumstances and the necessity for the gap to be relation between the actions of human beings
bridged by unsupported assertion and conjec- and the economic and political structures within
ture, and such linguistic practices as nominalism which they act. In the latter case, the theoretical
and obfuscation in the discussion of process. Why structure is scarcely brought to bear on the em-
is it necessary that our credulity be so strained to pirical work. Also, in this latter case, the identifi-
defend the orthodoxy of structuralism? cation of historical tendencies rather than the
The answer is that structural marxism does not specification of theoretical laws weakens sub-
always engage in conventional validation pro- stantially the domain of theoretical surety, and
cedures. When Engels presented his portrait of dismisses any claims for universalism. Which
the English working class, his posture was not that factors are more important in a given situation,
of the objective, detached investigator; in a and which less so, and how (with a structuralism
letter to Marx he described himself instead as the sceptical of surface appearances) are we to
prosecutor of the English middle class (Engels know the difference? What are we to make of
1958, p. xxi). Pronouncements of objectivity an historical situation in nineteenth-century Bri-
obscure the fact that theories may be subject to tain, the "most typical form" of capitalist society,
political not scientific validation, positing a spe- when as critical a theoretical proposition as the
cific class interest for judgment. "It is irrelevant," dichotomous class structure cannot be located?
writes Harvey (1973, p. 298), "to ask whether No wonder a strongly empirical radical analysis
concepts, categories and relationships are 'true' invites charges of theoretical heresy where there
is "little left to distinguish Marxism from a vacuous
or'false.' "Theirjustification, rather, rests with the
fact that they are "revolutionary theories which pluralism devoid of any predictive power"
are productive of change." Ultimately, the facts (Sayer 1975).
may become irrelevant, a methodological The status of borrowed theory is not as firm as
idiosyncrasy expressed by Lukdcs, who (quoting geographers have implied. There are many
Fichte) rejected the role of empirical refutation "marxisms," and the issues that separate them
with the phrase "so much the worse for the facts" are both complex and controversial. As we have
(Kolakowski 1978, Vol. 3, p. 265). Knowledge tried to show by the range of our citations, in-
and social interests are then self-consciously con- cluding E. P. Thompson, a marxist historian who
flated, as an arbitrary vantage point claims the has a fuller grasp of the complexity of the
status of universal scientific knowledge. But it is theoretical arguments and the negative verdict
precisely such "pure theory, wanting to derive of history on much marxist praxis, the debate is
everything from itself, that succumbs to unac- both internal as well as external to the marxist
knowledged external conditions and becomes traditions. The state of marxist theory refutes the
ideological" (Habermas 1971, p. 314).21 universalism and confidence that has been dis-
played in the geographic literature, and dis- integrity of the discipline. This is a matter not un-
counts the claim for an established, unitary familiar to geographers, for the nineteenth-
methodology or corpus of theory. century paradigm identifying a distinctive set of
As we saw, the holistic theoretical perspective, relations between people and the physical en-
which the marxist geographers have drawn vironment could also claim a distinctive disci-
from the Hegelian tradition, not only hinders plinary niche, but at the continual danger of
empirical analysis, but is also open to criticism as overspecifying the potency of the physical envi-
a purely theoretical formulation. Not only is an ronment while minimizing the role of human
ontological fallacy committed by reification and action and a discussion of process. There was, in
the attribution of causal primacy to abstractions, short, a purely practical and internal necessity for
but also these reifications, being derived from incomplete explanation in order to sustain the
nineteenth-century Hegelian idealism, have a identity of the discipline; there was, as A. F. Mar-
clear teleological character. Furthermore, as we tin (1951 ) lucidly put it for human geography, a
have shown, the form of the explanations is practical necessity for determinism.
both tautological and empirically untestable. The some tendencies are present in structural
The result is a mystification in explanation of how marxism. We noted earlier how economism has
real processes operate. been "the dominant basic form of Marxist
There is a final point to be made, though at theory," how a persistent marxian bias toward
this stage it is an argument that should be ad- economism is "a tendency that has yet to be
vanced tentatively, for its intricacies need further explained adequately" (Hirst 1977; Poster
exploration. At several points we have noted 1975a; Johnson 1978; Clarlke 1979; Thompson
theoretical parallels between structural marxism 1978b). The integrity of structural marxism is
and other holistic schools such as Durkheim's similarly predicated on theoretical overspecifi-
sociology and Kroeber's anthropology. In each cation, on a potent abstract structure which, it is
instance, the posing of a theoretical structure held, propels the course of history. The holistic
was accompanied by a passive view of human argument is characterized by determinacy and
beings and the reification of structure. It is perti- closure. But this thesis may be sustained only by
nent to heed the warning that "a purely struc- a certain mystification in the discussion of process
tural sociology is endemically in danger of reify- and by either an underdevelopment of or a
ing social phenomena. Even if it begins by demeaning neglect for empirical study. Too
modestly assigning to its constructs merely much attention to historical or geographical cir-
heuristic status, it all too frequently ends by con- cumstances transcends the terms of reference of
fusing its own conceptualizations with the laws of the theoretical structure and may lead to unac-
the universe" (Berger and Luckman 1967, p. 186). ceptable conclusions that challenge its integrity.
This much we have already demonstrated. The practical intent behind these tendencies in
But for Durkheim and Kroeber, the specifica- structural marxism needs to be exposed, for the
tion of an essential or superorganic structure had primary (if unrecognized) objective is less the
a practical as well as a theoretical purpose. Both explanation of the real world than the preserva-
followed Comte in ascribing to each discipline a tion of an abstract model of mechanism that is
distinctive set of phenomena to be studied. Both essentially intellectual. Paradoxically, the fun-
were concerned to protect the identity of their damental objective then becomes idealistic, the
own discipline and its associated set of problems defense of the mental categories of the author.
and explanations from reduction to another sci- These categories, which are projected onto the
ence, notably psychology. This practical objec- real world and given a life of their own, must be
demystified and referred back to the practical
tive consistently invited the risk of theoretical clo-
sure, of overly determined argument, and of a intent of the consciousness that created them. In
certain mystification in explanation as causal short, for structural marxists, ideology critique
power was assigned to a superorganic structure should be directed inward as well as outward.
while independent human action was di- In a letter sent to Marx in 1844, while he was
minished as it fell outside the terms of reference worlking on The Condition of the Working
of the social science in question. In other words, Class in England, Engels wrote: "We must start
mystification in the discussion of process was a out from empiricism and materialism, if our
result of a partial disciplinary perspective; but this thoughts and in particular our 'man' are to be
partial perspective was required to defend the something real. We must deduce the general
56 Duncan and Ley
from the particular and not from itself or out of 8. Marxist geographers have clearly been influ-
thin air d la Hegel" (cited in Stedman Jones enced by Marx in their organicism. In Grundrisse
as in Capital and elsewhere, Marx speaks of soci-
1977). Engels soon revised this assessment, and
ety as an organic system that subordinates all
the history of structural marxism has also been elements of society to itself and creates out of
one of continuous resistance to such a proposi- itself the organs that it laclks. See Marx (1973), p.
tion. Instead, the marxian authors have de- 278. Also relevant is Marx's great respect for Dar-
win's evolutionary theory. Kropotlkin, although not
clared (not demonstrated) their own authority;
a marxist himself, has influenced marxist geog-
by the stroke of a pen the subject has pro- raphers and also had an organicist view of society.
claimed itself aufgehoben. See Galois (1979).
9. Among the prominent organismic biologists or
vitalists, as they were called, were S. Haldane and
Hans Driesch. More recently, Paul Weiss and L.
Acknowledgments
Von Bertalanffy have argued for a holistic in-
terpretation of biology. However, since the late
The authors are grateful to John Agnew, Nancy Dun- nineteenth century this interpretation has never
can, David Evans, John Lowman, Geraldine Pratt, and held an appeal for mainline biologists. See Phillips
Gretchen Zdorkowski for their comments on an earlier (1976) for a critique of organismic biology. Also
draft of this paper. see Beclkner (1967).
10. For an exposition and defense of Marxist
structural-functionalism, see Cohen (1978).
11. See also Gregory (1979) and Kennedy (1979). It
Notes is significant also to note Harvey's earlier sympathy
for functionalism and systems analysis as positivist
1. This organicism often emerges in social science in models of explanation. Here as elsewhere there
the form of systems theory. is continuity rather than innovation in his radical
2. Some of the better known commentaries on formulations (Harvey 1969, Chaps. 22-23).
Marx's links to Hegel include Bernstein (1971), 12. Again there is continuity with positivist forms of
Avineri (1971), Ollman (1971), Hook (1962). geography in the 1960's, with their passive
3. The quotations from Marx are drawn from Marx model of man and their underpinning in neoclas-
(1973), pp. 459 and 276. sical economics. Compare Burton's (1963) obser-
4. It should be noted that Althusser much more con- vation that a "mechanistic flavour is present in
sistently devalues human activity than does Marx. much of the recent worlk by the 'quantifiers'....
5. Smith (1980) argues against Althusser's separa- The quantitative revolution is talking us baclk much
tion of natural and social science, and the separa- closer to environmental determinism."
tion of the "real object" from the "object of 13. For this view from other perspectives, see Gould-
knowledge." He does not, however, focus on Alt- ner (1979) and Parlkin (1980).
husser's use of reified wholes as we do here. 14. See, for example, Bechhofer and Elliott (1976).
6. Idealism, in its philosophical sense, is the view that 15. Although there were ten shopkeepers on Man-
mind and ideas are simply not reducible to mate- chester's first town council in 1838 (Briggs 1965, p.
rial things and processes. It is also opposed to the 105).
commonsense view that material things exist in- 16. "Marx's terminology is careless" (Giddens 1973,
dependently of their being perceived. Transcen- pp. 27, 28). Ollman is more severe: "With all due
dental idealism such as Hegel's is the view that allowance made for loose word usage, however,
there is a transcendent totality of which material Marx cannot escape the more serious accusation
entities including man are manifestations. These, of having a litter of standards for class member-
in very brief and simplified form, are the standard ship and of changing them without prior warning"
definitions used in philosophy. However, social (Ollman 1968).
scientists, the marxist geographers included, 17. See, for example, Moorehouse (1978); Stedman
sometimes use the term idealism to refer to a Jones (1976); the debate running through His-
view that overemphasizes the ideas and inten- tory Workshop in 1978-79; and, in particular,
tions of individuals, and that sometimes leads to Thompson's (1 978b) defense of historical method
what is termed "conspiracy theory" in political and assault on structural theoreticism.
analysis. In this paper we use the term in its 18. For a historical analysis offering different conclu-
philosophical sense. Marxist geographers are guilty sions, see Holdsworth (1980).
of philosophical idealism generally and transcen- 19. For a bibliography of the earlier studies, see
dental idealism especially, because they give on- Johnston (1977).
tological status and causal primacy to transcen- 20. For a review of this debate, see Williams (1978);
dental, supra-individual entities, while accusing a broader view of the politics of consumption ap-
anyone who sees individual reasons as possible pears in Cawson (1977).
causes of events of idealism. 21. Of course, historical and political praxes do not
7. Although Castells is an urban sociologist, he is in- validate the theoretical prospectus either, so that
cluded here because his work has an important structural marxism is deeply flowed even on its
spatial component and because he has influ- own grounds for validation. Indeed, "the greatest
enced a number of geographers. ideal obstacle has been the caricature of socialism
presented by the fact of socialist states" Clorke, S. 1979. Socialist humanism and the critique of
(Thompson 1978b, p. 169). economism. History Workshop 8:137-56.
Clovol, P. 1977. Le marxisme et lespace. L'Espace
Goographique 6:145-64.
Cohen, G. 1978. Karl Marx's theory of history: a
defence. Oxford: Clarendon Press.
References Cox, K. 1981. Capitalism and conflict around the
communal living space. In Urbanization and
Agnew, J. A., and Duncon, J. S. 1981. The transfer of urban planning in capitalist society, eds. M.
ideas into Anglo-American human geography. Dear and A. Scott, pp. 431 -55. New York: Me-
Progress in Human Geography 5:42-57. thuen.
Alonso, W. 1965. Location and land use. Cam- Duncan, J. S. 1980. The superorganic in American
bridge: Harvard University Press. cultural geography. Annals of the Association of
Althusser, L. 1969. For Marx. Harmondsworth, U.K.: American Geographers 70:181 -98.
Penguin. Dupr6, L. 1974. Recent literature on Marx and marx-
. 1972. Politics and history: Montesquieu, ism. Journal of the History of Ideas 35:703-14.
Rousseau, Hegel and Marx. London: New Left Elliott, B., and McCrone, D. 1979. Power and protest in
Books. the city. Unpublished paper, Department of
Althusser, L., and Balibor, E. 1970. Reading Capital. Sociology, University of Edinburgh.
London: New Left Books. Engels, F. 1958. The condition of the working class
Anderson, J. 1973. Ideology in geography: an intro- in England. Eds. W. Henderson and W. Chaloner.
duction. Antipode 5:1 -6. Oxford: Blaclkwell.
Avineri, S. 1971. The social and political thought of Feuerboch, L. 1957. The essence of Christianity.
Karl Marx. London: Cambridge University Press. Eds. E. G. Waring and F. W. Strothmann. New
Baudrillord, J. 1975. The mirror of production. St. York: F. Ungar.
Louis: Telos Press. Galois, B. 1979. Ideology and the idea of nature: the
Bechhofer, F., and Elliott, B. 1976. Persistence and case of Peter Kropotlkin. Antipode 8:1 -16.
change: the petite bourgeoisie in industrial soci- Gans, H. 1967. The Levittowners. New York: Pan-
ety. European Journal of Sociology 17:74-99. theon Books.
Beckner, M. 0. 1967. Organismic biology. In The en- Giddens, A. 1973. The class structure of the ad-
cyclopedia of philosophy, ed. P. Edwards, vols. vanced societies. London: Hutchinson.
5 and 6, pp. 349-551. New York: Macmillan. . 1979. Central problems in social theory.
Berger, P., and Luckmon, T. 1967. The social con- Berkeley: University of California Press.
struction of reality. New York: Doubleday Glucksmonn, M. 1974. Structuralist analysis in
Anchor. contemporary social thought. London: Rout-
Berger, P., and Pullberg, S. 1964-65. Reification and ledge and Kegan Paul.
the sociological critique of consciousness. History Gouldner, A. 1979. The future of intellectuals and
and Theory 4:196-211. the rise of the new class. New York: Seabury.
Bernstein, R. 1971. Praxis and action: contemporary . 1980. The two marxisms. New York: Sea-
philosophies of human action. Philadelphia: bury.
University of Pennsylvania Press. Gray, F. 1976. Radical geography and the study of
Blout, J. M. 1974. The ghetto as an internal neo- education. Antipode 8:38-44.
colony. Antipode 6:37-41. Gregory, D. 1979. The ideology of control: systems
-. 1975. Imperialism: the marxist theory and its analysis and geography. Paper presented to the
evolution. Antipode 7:1 -19. Anglo-French Symposium on Geography and
Briggs, A. 1965. Victorian cities. New York: Harper Ideology, Cambridge.
and Row. . 1981. Human agency and human geogra-
Burowoy, M. 1978. Contemporary currents in marxist phy. Transactions, Institute of British Geog-
theory. The American Sociologist 13:50-64. raphers N.S. 6:1-18.
Burton, I. 1963. The quantitative revolution and Habermos, J. 1971. Knowledge and human inter-
theoretical geography. Canadian Geographer ests. Boston: Beacon Press.
7:151 -62. Hall, S. 1977. The "political" and the "economic" in
Castells, M. 1976. Theoretical propositions for an ex- Marx's theory of classes. In Class and class
perimental study of urban social movements. In structure, ed. A. Hunt, pp. 15-60. London:
Urban sociology: critical essays, ed. C. Lawrence and Wishart.
Piclkvance, pp. 147-73. London: Tavistock. Harnilton, P. 1974. Knowledge and social structure.
. 1977. The urban question. London: Edward London: Routledge and Kegan Paul.
Arnold. Harloe, M. 1977. Introduction. In Captive cities, ed.
1978. City, class and power. London: Mac- M. Harloe, pp. 1-47. London: Wiley.
millan. Harvey, D. 1969. Explanation in geography. Lon-
Castells, M., and de Ipola, E. 1976. Epistemological don: Arnold.
practice and the social sciences. Economy and -. 1972. Revolutionary and counter-revolution-
Society 5:111 -44. ary theory in geography and the problem of
Cawson, A. 1977. Pluralism, corporatism, and the role ghetto formation. Antipode 5:1 -6.
of the state. Government and Opposition . 1973. Social justice and the city. London:
13:178-98. Arnold.
58 Duncan and Ley
. 1975a. Class structure in a capitalist society Kennedy, B. 1979. A naughty world. Transactions,
and the theory of residential differentiation. In Institute of British Geographers N.S. 4:550-58.
Processes in physical and human geography, Kolokowski, L. 1978. Main currents of marxism, 3
eds. R. Peel et al., pp. 354-69. London: vols. Oxford: Clarendon Press.
Heinemann. Kovel, J. 1976. The marxist view of man and
- . 1975b. The geography of capitalist accumu- psychoanalysis. Social Research 43:220-45.
lation: a reconstruction of marxian theory. Lambert, J., Paris, C., and Blockoby, B. 1978. Hous-
Antipode 7:9-21.
ing policy and the state. London: Macmillan.
. 1975c. The political economy of urbanization
Lee, R. 1979. The economic basis of social problems
in advanced capitalist societies: the case of the
in the city. In Social problems and the city: geo-
United states. In The social economy of cities,
graphical perspectives, eds. D. Herbert and D.
eds. G. Gappert and H. Rose, pp. 119-63. Bev-
Smith, pp. 47-62. Oxford: Oxford University
erly Hills: Sage.
Press.
. 1975d. Union of socialist geographers-
Ley, D. 1978. Social geography and social action. In
seminar in marxist geography. Antipode 7:
54-86. Humanistic geography, eds., D. Ley and M.
. 1976. The marxian theory of the state.
Samuels, pp. 41 -57. Chicago: Maaroufa.
Antipode 8:80-89. . 1979. Ideology, theory, and empirical study
-. 1978a. Labor, capital and class struggle in marxian approaches to urban geography.
around the built environment in advanced Paper presented to the Anglo-French Symposium
capitalist societies. In Urbanisation and conflict on Geography and Ideology, Cambridge.
in market societies, ed. K. R. Cox, pp. 9-37. Ley, D., and Mercer, J. 1980. Locational conflict and
Chicago: Maaroufa. the politics of consumption. Economic Geogra-
-. 1978b. The urban process under capitalism: a phy 56:89-109.
framework for analysis. International Journal of Lukes, S. 1973. Individualism. Oxford: Basil
Urban and Regional Research 2:101-31. Blaclkwell.
. 1979. Myth and monument. Annals of the . 1977. Essays in social theory. London: Mac-
Association of American Geographers 69: millan.
362-81. Marcus, S. 1974. Engels, Manchester, and the
Hegel, G. W. F. 1953. Reason in history: a general working class. New York: Random House.
introduction to the philosophy of history. Martin, A. 1951. The necessity for determinism.
Translated with an introductic., by R. S. Hartmon. Transactions, Institute of British Geographers
New York: The Liberal Arts Press. 17:1-11.
Hirst, P. 1977. Economic classes and politics. In Class Marx, K. 1954. The eighteenth brumaire of Louis
and class structure, ed. A. Hunt, pp. 125-54. Bonaparte. Moscow: Progress Publishers.
London: Lawrence and Wishart. . 1964. Early writings. Translated and edited
. 1979. The necessity of theory. Economy and by T. Bottomore. New York: McGraw Hill.
Society 8:417-45. . 1967a. Capital: a critique of political econ-
Holdsworth, D. 1980. Bread and roses: the struggle for omy, 3 vols. Ed. F. Engels. New York: Interna-
home in Vancouver and Cape Breton. Unpub- tional Press.
lished paper, Department of Geography, Univer- * 1967b. The economic and philosophical
sity of Toronto. manuscripts of 1844. In Writings of the young
Hook, S. 1962. From Hegel to Marx: studies in the Karl Marx on philosophy and society, eds. L. D.
intellectual development of Karl Marx. Ann Easton and K. H. Guddat, pp. 283-337. New
Arbor: University of Michigan Press. York: Doubleday.
. 1980. Spectral marxism. The American -. 1972. Theses on Feuerbach. In The Marx-
Scholar 49:250 et seq. Engels reader, ed. R. C. Tucker, pp. 107-9.
Hoyt, H.- 1933. One hundred years of land values in New York: W. W. Norton.
Chicago. Chicago: University of Chicago Press. . 1973. Grundrisse. London: Pelican.
Israel, J. 1971. Alienation from Marx to modern Massey, D., and Catolono, A. 1978. Capital and land:
sociology. Boston: Allyn and Bacon. landownership by capital in Great Britain. London:
Johnson, R. 1978. Edward Thompson, Eugene Arnold.
Genovese, and socialist-humanist history. History Moorehouse, H. 1978. History, sociology and the
Workshop 6:79-100. quiescence of the British worlking class: a reply t
Johnston, R. 1977. Urban geography: city structures. Reid. Social History 4:481 -90.
Progress in Human Geography 1: 118-29. Nield, K. 1980. A symptomatic dispute? Notes on the
Jones, G. Stedmon. 1976. From historical sociology to relation between marxian theory and historical
theoretical history. British Journal of Sociology practice in Britain. Social Research 47:479-506.
27:295-305. Olmaon, B. 1968. Marx's use of "class." American
. 1977. Engels and the genesis of marxism. Journal of Sociology 73:573-80.
New Left Review 106:79-104. . 1971. Alienation: Marx's conception of man
-. 1979. History and theory. History Workshop in capitalist society. Cambridge: Cambridge
8:198-202. University Press.
Keat, R., and Urry, J. 1975. Social theory as G.
Olsson, science.
1972. On reason and reasoning. Antipode
London: Routledge and Kegan Paul. 4:26-31.
O'Neill, J. 1973. Modes of individualism and collec- velopment. In The problem of the cities and
tivism. London: Heinemann Educational Books. towns, ed. G. Greer, pp. 6-17. Cambridge:
Pahl, R. 1976. Whose city? London: Penguin. Harvard University Press.
. 1979. Socio-political factors in resource allo- Santos, M. 1975. Space and domination: a marxist
cation. In Social problems and the city, eds. D. approach. International Social Science Journal
Herbert and D. Smith, pp. 33-46. Oxford: Ox- 27:346-63.
ford University Press. . 1977. Society and space: social formation as
Parkin, F. 1979. Marxism and class theory: a theory and method. Antipode 9:3-13.
bourgeois critique. London: Tavistoclk. Saunders, P. 1978. Domestic property and social class.
. 1980. The academicizing of marxism. Dis- International Journal of Urban and Regional
sent 27:172-80. Research 2:233-51.
Peet, R. 1969. The spatial expansion of commercial Sayer, D. 1975. Method and dogma in historical ma-
agriculture in the nineteenth century; a von terialism. Sociological Review 23:779-810.
Thuinen interpretation. Economic Geography Schrog, C. 0. 1975. Proxis and nature: conflicting
45:283-301. models in the science of man. Journal of the
. 1975. Inequality and poverty: a marxist- British Society for Phenomenology 6:23-31.
geographic theory. Annals of the Association of Short, J. R. 1976. Social systems and spatial, patterns.
American Geographers 65:564-71. Antipode 8:77-83.
. 1977a. The development of radical geogra- Skidmore, W. 1979. Theoretical thinking in sociol-
phy in the United States. Progress in Human Ge- ogy. Cambridge: Cambridge University Press.
ography 1:240-63. Smith, N. 1979a. Gentrification and capital: practice
. 1977b. Interaction of the environmental and and ideology in Society Hill. Antipode
spatial contradictions during late capitalism. 11:24-35.
Antipode 9: inside front cover. . 1979b. Toward a theory of gentrification.
. 1978. The dialectics of radical geography: a Journal of American Planning Association
reply to Gordon Clarlk and Michael Dear. The 45:538-48.
Professional Geographer 30:360-64. . 1980. Symptomatic silence in Althusser: the
. 1979. Societal contradictions and marxist concept of nature and the unity of science. Sci-
theory. Annals of the Association of American ence and Society 44: 58-81.
Geographers 69:164-69. Spencer, M. 1977. History and sociology: an analysis
Peet, R., and Lyons, J. V. 1981. Marxism: dialectical of Weber's "The City." Sociology 11:507-25.
materialism, social formation and the geographic Thompson E. P. 1978a. Eighteenth century English so-
relations. In Themes in geographic thought, eds. ciety: class struggle without class. Social History
M. E. Harvey and B. P. Holly, pp. 187-205. Lon- 3:133-65.
don: Croom Helm. . 1978b. The poverty of theory and other es-
Phillips, D. C. 1976. Holistic thought in social sci- says. London: Merlin Press.
ence. Stanford: Stanford University Press. Timms, D. 1971. The urban mosaic. Cambridge:
Pickvance, C. 1976a. Housing, reproduction of capi- Cambridge University Press.
tal, and reproduction of labor power: some re- Tristram, R. 1975. Ontology and theory: a comment
cent French worlk. Antipode 8:58-68. on Marx's analysis of some of the problems.
, ed. 1976b. Urban sociology: critical essays. Sociological Review 23:759-71.
London: Tavistoclk. Walker, R. A. 1978. The transformation of urban
Poster, M. 1975a. Existential marxism in postwar structure in the nineteenth century and the begin-
France. Princeton: Princeton University Press. nings of suburbanization. In Urbanization and
. 1975b. Translator's introduction. In The mirror conflict in market societies, ed. K. R. Cox, pp.
of production, J. Baudrillard, pp. 1 - 15. St. Louis: 165-212. Chicago: Maaroufa Press.
Telos Press. Ward, D. 1975. Victorian cities: how modern? Journal
Poulantzas, N. 1979. State, power, and socialism. of Historical Geography 1: 135 - 51.
New Yorlk: Schocden. 1976. The Victorian slum: an enduring myth?
Preteceille, E. 1976. Urban planning: the contradic- Annals of the Association of American Geog-
tions of capitalist urbanization. Antipode raphers 66:323-36.
8:73-74. . 1980. Environs and neighbours in the "Two
Quinn, J. 1940. The Burgess zonal hypothesis and its Nations." Journal of Historical Geography
critics. American Sociological Review 5:210-18. 6:133-62.
Roweis, S., and Scott, A. 1978. The urban land ques- Williams, P. 1978. Urban managerialism: a concept of
tion. In Urbanization and conflict in market relevance? Area 10:236-40.
societies, ed. K. Cox, pp. 38-75. Chicago: Wisner, B. 1978. Does radical geography lack an ap-
Maaroufa Press. proach to environmental relations? Antipode
10:84-95.
Ryon, A. 1970. The philosophy of the social sci-
Wright, E. 1976. Class boundaries in advanced
ences. London: Macmillan.
capitalist societies. New Left Review 98:3-41.
, ed. 1973. The philosophy of social expla- Wrong, D. 1961. The oversocialized conception of
nation. London: Oxford University Press. man in modern sociology. American Sociologi-
Saey, P. 1978. Marx and the students of space. L'Es- cal Review 26:183-93.
pace Geographique 7:15-25. Zehner, R. 1977. Indicators of the quality of life in
Samuelson, P. 1942. The business cycle and urban de- new communities. Cambridge, Mass.: Ballinger.

DUNCAN LEY Critic Marxism 1982

Uploaded by

Copyright:

Available Formats

DUNCAN LEY Critic Marxism 1982

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DUNCAN LEY Critic Marxism 1982

Uploaded by

Copyright:

Available Formats

Structural Marxism and Human Geography: A Critical Assessment

Author(s): James Duncan and David Ley

Structural Marxism and Human Geography:

James Duncan and David Ley

Department of Geography, University of British Columbia, Vancouver, British Columbia V6T 1 W5

odological, theoretical, and practical com- economics-though to suggest thot questions of

idealism. Other examples of the Hegelion portant elements of Oilman's interpretation of

This is an interesting statement in part because it

necessary beyond establishing that people re-

You might also like