an institution-free society


Despite occasional flirtations with rational choice theory, I am not (and never will be) a methodological individualist.  One reason I’ve given for avoiding this kind of reductionism can be summarized in a recent article by Geoffrey Hodgson in Organization Studies (“Institutions and individuals: Interaction and evolution”).  Disclaimer: I haven’t read the entire article and so I can’t vouch for Hodgson’s conclusions.

The narrow methodological individualist has a problem of infinite regress: attempts to explain each emergent layer of institutions always rely on previous institutions and rules. These in turn have to be explained….All theories must first build from elements which are taken as given. However, the particular problems identified here undermine any claim that the explanation of the emergence of institutions can start from some kind of institution-free ensemble of (rational) iindividuals in which there is supposedly no rule or institution to be explained. Consequently, the project to explain the emergence of institutions on the basis of given individuals runs into difficulties, particularly with regard to the conceptualization of the initial state of nature from which institutions are supposed to emerge.

As I read this statement now, I wonder if Hodgson’s view of methodological individualists is correct.  Is it too simplistic to say that individualists assume the conditions of an institution-free world?   Many game theorists now consider how institutions evolve in the context of existing institutions.  For example, Avner Grief’s historical and comparative institutional analysis is based on game theoretic choices given certain exogenous institutional constraints (e.g. a belief system).  Organizations interact with the environment, altering the existing rules to create new institutions that generate certain strategic advantages.

How correct is Hodgson’s perspective?  Does methodological individualism lead to an infinite regress to assumptions of an institution-free world or is this just a convenient strawman?

Written by brayden king

February 15, 2007 at 12:04 am

Posted in brayden, economics

8 Responses

Subscribe to comments with RSS.

  1. I haven’t read the article, but the infinite regress assumption is as much a problem for structuralistas as it is for meth-(ind)-heads (as our friend Satoshi Kanazawa reminds us). In fact, I would argue that the infinite regress assumption is much more of a problem for any ontological structuralist account than from an ontological individualist story, since structures cannot (by definition) be foundational entities (while individuals may be but not always); they must be generated from some other set of simpler components (individuals may be this component, but not necessarily). Thus any generative structuralist stance truly worth its salt cannot be satisfied with pointing the finger at the meth-heads for having the logical difficulty of infinite regress. It must point that finger right back at itself.



    February 15, 2007 at 1:16 am

  2. MI and infinite regress. If we are trying explain institutions, the natural/logical place to start is with lower-level (composing) antecedents – MI – rather than take the existence of collective structures (institutions, orgs etc) for granted (or, rather than simply cite historicism/evolution, which creates a far worse regress problem where [for example] institutions create institutions create institutions; but, where do institutions, structures, networks etc originate from in the first place is a more fundamental question, which points squarely to lower levels).

    Now, of course, we are idealizing/theorizing here, so some purposeful “mistakes” will get made – below Coleman broadly on the matter (though, the best answer to this question is in Coleman, 1990: specifically chapter 1):

    ‘‘In this paper, I will proceed in precisely the opposite fashion to that taken by the advocates of homo sociologicus. I will make an opposite error, but one which may prove more fruitful. I want to begin the development of a theory of collective decisions, and in so doing I will start with an image of man as wholly free: unsocialized, entirely self-interested, not constrained by norms of a system, but only rationally calculating to further his own interest. This is much the image of man held by economists, and with it the economists have answered one part of Hobbes’s question: how is it that although the men who make it up are wholly self-interested, the economic system can operate with one man’s actions benefiting others. It was the genius of Adam Smith to pose an answer to this part of Hobbes’s question’’ (Coleman, 1964: 167).

    In short, respectfully, Hodgson is wrong.



    February 15, 2007 at 2:52 am

  3. Thanks for answering my question Teppo. Apparently you (and Coleman) are the true and living MI of which Hodgson wrote.

    Of course, by throwing Coleman out there you’re not actually demonstrating that Hodgson is wrong. You’re only showing that Coleman preferred the institution-free scenario to the other. You’re basically agreeing with Hodgson. The difference between the two of you is that one prefers institution-free theorizing while the other does not. It’s hard to argue that one person is right and the other is wrong without generating some criteria to evaluate the theories that are not based on personal preferences.

    My bias is towards a more historical-evolutionary model. I find much more value in Grief’s perspective (but again, I’m not saying it’s more logical, but I do find it more useful). Assume institutions exist and then theorize how individuals make choices and strategize within that world. I find it more interesting because, well, it seems to give us more leverage in understanding how humans actually behave and change organizations and other institutions in the real world than does a theory that begins with the assumptions of an institution-free society. I can empirically observe one and I can’t observe the other. Makes a nice story though.



    February 15, 2007 at 5:39 am

  4. I think a “choice-within-constraints”-type approach (Ingram and Clay) certainly has a nice feel to it, or more of a Northian or public choice-type approach (I believe these generally work better in resolving this problem than simply discarding levels in Giddensian fashion, or alternatively, taking higher levels for granted).

    In terms of preferences (institution-free vs. institution-laden explanation), I think there is more at stake. We can perhaps take a multi-lense approach and say that both elucidate and help us understand, and that in fact seems to be the approach disciplines (in part) settle on, or, we can simply just be pragmatic and say that the approach depends on the problem/question at hand; but, there are many, many times when the two fundamentally clash as well, and no amount “can’t-we-all-just-get-along,” or hand-waving (or, quoting – like I just did :) helps reconcile things.

    (For an institutional angle on this, Selznick makes (roughly) the same point in his 1996 ASQ piece, in response to problems with higher-level theories in neo-institutional theory).



    February 15, 2007 at 5:47 am

  5. Ok, I read the paper, now I can trash it.

    To begin with, one problem with the Hodgson’s paper is there is more substance in the brief exchange between Brayden and Teppo that precedes this comment than in Hodgson’s entire paper. Hodgson’s paper belongs to a disturbing genre of post-Giddensian British social theory which is so concerned with scholastic matters of definition and conceptual refinement that it is seldom able to make any clear analytical points. Consider for instance the following:

    Before these arguments are developed, some terminological issues require attention. What is the difference between social structures, institutions, conventions and organizations? As elaborated elsewhere (Hodgson 2006), these terms are used in the following way:

    • Social structures include all sets of social relations, including the episodic and those without rules, as well as social institutions.
    • Institutions are systems of established and embedded social rules that structure social interactions.
    • Rules in this context are understood as socially transmitted and customary normative injunctions or immanently normative dispositions that in circumstances X do Y.
    • Conventions are particular instances of institutional rules.
    • Organizations are special institutions that involve (a) criteria to establish their boundaries
    and to distinguish their members from non-members, (b) principles of sovereignty concerning who is in charge and (c) chains of command delineating responsibilities within the organization (96).

    Rivers of ink have been spilled trying to sort out through these terms in the literature, yet the author disposes of the thorny issues in a quick paragraph. The problem with these “definitions” is that they are either wrong or entirely vacuous. Is it helpful to think of organizations as “special institutions”? Are all organizations endowed with “principles of sovereignty” and “chains of command”? What does Hodgson mean by the the statement that institutions are “established and embedded”? In what? If the term social structure refers to “all sets of social relations” then what is not social structure?

    One could go on, but why bother? The rest of Hodgson’s paper reads like a well crafted graduate student paper in a Brit-soc-theory seminar. The problem with these papers is that they make me roll my eyes so much that I get a headache when reading them (which is why I tend to keep the Advil near when going through the table of contents of Sociological Review or the BJS; yes I am sucker, but I keep hoping that I will run into something good). The model of theoretical analysis is what one could call the “Goldie-Locks strategy.” This theory is too individualist; this theory is too collectivist; this theory is just right. This theory ignores agency; this theory ignores structure; this theory is just right; this theory ignores power; this theory reduces everything to power relations; this theory is just right…and so on, and so on. Giddens established this model in a series of books in the 1970s and 1980s (i.e. New Rules [1976], Central Problems [1979], Constitution [1984], etc.). It is no surprise that younger British scholars have turned the Goldie-Locks model against the master: pre-Giddens (British) social theory radically separated Agency and Structure; Giddens radically melds agency and structure; critical realism (in allowing for analytical separation and interaction between the two) is just right; give me a freakin’ break!

    The rest of the paper is pretty much Goldie-Locks against various straw-man versions of “collectivism” and MI. Unfortunately the payoff is so trite, that one wonders why go through the whole excercise in the first place. The author concludes:

    What is required is a framework within which the transformation of both individuals and structures can be explained. This approach must involve explanations of possible causal interaction and reconstitution, from both individual to structure and from structure to individual. This would mean an explanation of the evolution of individual purposes and beliefs, as well as an explanation of the evolution of structures. Preferences or purposes would be endogenously formed. Their co-evolution must be examined, without conflating one into the other. Such an evolutionary analysis provides the means by which social theory may escape from its unsustainable dichotomies and make further progress (106).

    Where’s my copy of Foundations of Social Theory? Is there an entry for “reinventing the wheel” in the index of this 1000-page book?



    February 15, 2007 at 10:42 am

  6. Yep, that last paragraph is sufficiently vague as to envelope about 3/4 of all organizational sociology done in the last decade.

    After reading Omar’s thrashing of the paper, I’m relieved that I didn’t endorse it at the beginning of the post.



    February 15, 2007 at 2:32 pm

  7. […] a meeting…) But I wanted to pass along this quote, in which Omar Lizardo characterises a very common academic argumentative strategy almost too precisely for comfort: The model of theoretical analysis is what one could call the […]


  8. Randy Collins has anice essay about the infinite regress problem of methodological individualism in his book _Sociological Insight: AN Introduction to Nonobvious Sociology._ Its called “Nonrational foundations to Sociology.” I remember it as being very lucid and effective in an undergraduate class I taight awhile back.



    February 18, 2007 at 2:55 am

Comments are closed.

%d bloggers like this: