Science and the Structure of Explanation in Evolutionary Theory and Historical Materialism
In a comment on my previous post Pam asks whether Darwin's theory of evolution might be subject to the same sort of worries regarding its scientific status (e.g. those raised by Popper) as Marx's theory of history. She writes:
With regard to Pam's suggestion that perhaps "the theory of evolution is largely true simply in virtue of the fact that whatever organisms happened to survive in the past were defined as the fittest," it is important to get clear on exactly what the theory entails. It's not quite right that contemporary evolutionary theory, which incorporates contemporary genetic theory, defines the organisms that survive as the fittest. Rather, the traits that survive within populations are thought to be those that make the organism best suited to survive in the environment in which the population lives. Furthermore, the fact that those traits survive is explained by the fact that they are adaptive for the population in the relevant environment. Adaptive traits are selected for, and the fact that they are selected for helps explain why, within populations as a whole, some traits survive and others don't.
Levine and Sober, in their article "What's Historical About Historical Materialism," suggest that there is a similar structure to historical materialist explanations: "...there will be selection for that set of production relations which is optimal for further developing productive forces." The idea is that at various points in the history of the development of the productive forces changes in production relations were necessary in order to allow the productive forces to continue developing, and that when that is the case the production relations do in fact tend to change so as to allow the further development of the forces. Changes in the production relations are explained by the fact that such changes were necessary in order to allow the productive forces to continue to develop. This, of course, requires the assumption of what Cohen calls the "Development Thesis," which is that the productive forces tend to develop throughout history.
In another comment on my previous post, Gay Species suggested that Marx's theory of history ought to be rejected because it is teleological; but this is incorrect, since Marx's theory can be understood to have roughly the same explanatory structure as evolutionary theory. Just as particular traits are not selected for because their being so moves evolution toward some end that it is aiming at, but rather because, at the time at which they are selected for, they are adaptive for the population relative to its environment, production relations are not, according to historical materialism, selected for because they move history toward some end that it is in any sense "aiming at" (though this is an understandable misinterpretation of Marx given the fact that he refers to the transition to communist relations of production as marking "the end of history"); rather they are selected for because, at the time at which they arise, they are optimal for further developing the productive forces.
Of course there are important differences between the explanatory structures of evolutionary theory and historical materialism as well, and these differences, according to Levine and Sober, give us reasons to be much more skeptical of both historical materialism's truth and its claim to scientific status. For example, evolutionary theory explains macro changes in terms of micro changes, whereas historical materialism explains micro changes by means of explaining macro changes. That is, large-scale changes in the frequencies of particular genes within a population over time are explained by evolutionary theory by the fact that early on, those members of the population with adaptive traits tended to survive longer and procreate more, whereas those lacking such traits tended to die sooner and procreate less. The result of this is that over time, the frequency of adaptive traits within the population increases, and the frequency of non-adaptive traits decreases. The increase in the frequency of adaptive traits (a macro-evolutionary change) is merely a fortuitous consequence (though a predictable one) of the effects of the environment on the procreative success of individual organisms. The same holds for even larger scale evolutionary changes such as speciation. Historical materialism, on the other hand, does not explain changes in the mode of production (e.g. the change from feudalist to capitalist relations of production) as merely the fortuitous consequence of many isolated changes in the relations of production in which individuals stand. Rather, the fact that the relations in which individuals stand change is explained by the fact that the mode of production itself changed, and changes in the mode of production are in turn explained by the fact that the old relations of production had become fetters on the development of the productive forces (and, of course, the Development Thesis). Explanitorily speaking, large scale changes are prior to small scale changes in historical materialism.
Now the explanatory priority of large scale changes within historical materialism is one aspect of the theory that makes it suspect in a way that evolutionary theory is not. For one might object, couldn't some small scale changes in the relations of production occur without the mode of production undergoing a thorough change. Now of course in a purely logical sense, this is entirely possible, but the mere logical possibility of such changes does not itself undermine historical materialism, since it is a theory about actual historical development and change. But still, we might think there are actual historical examples of such changes, including perhaps the growth of the French burghers during late feudalism (an example that Marx himself discussed).
This is a challenge that historical materialism may not be able to meet. But it is important to recognize that if we think that it is not possible to explain small scale historical changes in terms of large scale changes, then we are necessarily led to the conclusion that there can be no systematic theory of historical development at all. We cannot construct a theory of history that systematically explains large scale changes as the results of small scale changes, because there is no mechanism analogous to natural selection to explain small scale historical changes. And it seems to me that it is even more unlikely that there is a mechanism of this sort than that some theory with the explanatory structure (though perhaps not the content) of historical materialism is true. So, if we reject the explanatory structure of historical materialism, then we are left with the view that, as Levine and Sober put it, "history is irreducibly heterogeneous...and therefore ineluctably atheoretical." This is the view that most people accept, and perhaps it is right. If so, then history simply isn't a proper subject for scientific study, and there can be no systematic theory of history. But as Levine and Sober rightly point out, "whether history admits of a general theory should be discovered, not settled by speculative fiat." Historical materialism may very well be false; but if it is, it cannot be proven so merely by citing the apparent heterogeneity of explanations of individual historical events.
In a philosophy of science course I took, we spent a great deal of time discussing testability & falsifiability demarcation conditions for science. Not too surprisingly, Marx's theory of history came up as an example of a pseudoscientific practice that fails those conditions...I always struggled with one particular possible counterexample to testability/falsifiability: the purportedly scientific theory of evolution. Do you think the same argument you made about Marx can be made about evolution? (That is, the theory of evolution is largely true simply in virtue of the fact that whatever organisms happened to survive in the past were defined as the fittest).This is an interesting question, and there are several important issues to distinguish. But first, let me point out that it seems to me that the theory of evolution is in fact a falsifiable theory. If the remains of an organism of a particular species were found and dated to a historical period in which evolutionary theory cannot account for its presence (e.g. if a homo sapiens skeleton were dated to a million years ago), then the theory would be falsified. In the same way, historical materialism is a falsifiable theory. If, for example, we observed the productive forces continuing to develop in the current capitalist period, and at the same time observed the return of feudalist relations of production, historical materialism would be conclusively falsified. So Popper was wrong to claim that historical materialism is fundamentally unfalsifiable. Still, it may be that it is consistent with too many possible observations to be adequately tested by observation and examination of past historical events. If this is the case then historical materialism ought to be rejected as too vague in its implications, rather than as fundamentally unfalsifiable.
With regard to Pam's suggestion that perhaps "the theory of evolution is largely true simply in virtue of the fact that whatever organisms happened to survive in the past were defined as the fittest," it is important to get clear on exactly what the theory entails. It's not quite right that contemporary evolutionary theory, which incorporates contemporary genetic theory, defines the organisms that survive as the fittest. Rather, the traits that survive within populations are thought to be those that make the organism best suited to survive in the environment in which the population lives. Furthermore, the fact that those traits survive is explained by the fact that they are adaptive for the population in the relevant environment. Adaptive traits are selected for, and the fact that they are selected for helps explain why, within populations as a whole, some traits survive and others don't.
Levine and Sober, in their article "What's Historical About Historical Materialism," suggest that there is a similar structure to historical materialist explanations: "...there will be selection for that set of production relations which is optimal for further developing productive forces." The idea is that at various points in the history of the development of the productive forces changes in production relations were necessary in order to allow the productive forces to continue developing, and that when that is the case the production relations do in fact tend to change so as to allow the further development of the forces. Changes in the production relations are explained by the fact that such changes were necessary in order to allow the productive forces to continue to develop. This, of course, requires the assumption of what Cohen calls the "Development Thesis," which is that the productive forces tend to develop throughout history.
In another comment on my previous post, Gay Species suggested that Marx's theory of history ought to be rejected because it is teleological; but this is incorrect, since Marx's theory can be understood to have roughly the same explanatory structure as evolutionary theory. Just as particular traits are not selected for because their being so moves evolution toward some end that it is aiming at, but rather because, at the time at which they are selected for, they are adaptive for the population relative to its environment, production relations are not, according to historical materialism, selected for because they move history toward some end that it is in any sense "aiming at" (though this is an understandable misinterpretation of Marx given the fact that he refers to the transition to communist relations of production as marking "the end of history"); rather they are selected for because, at the time at which they arise, they are optimal for further developing the productive forces.
Of course there are important differences between the explanatory structures of evolutionary theory and historical materialism as well, and these differences, according to Levine and Sober, give us reasons to be much more skeptical of both historical materialism's truth and its claim to scientific status. For example, evolutionary theory explains macro changes in terms of micro changes, whereas historical materialism explains micro changes by means of explaining macro changes. That is, large-scale changes in the frequencies of particular genes within a population over time are explained by evolutionary theory by the fact that early on, those members of the population with adaptive traits tended to survive longer and procreate more, whereas those lacking such traits tended to die sooner and procreate less. The result of this is that over time, the frequency of adaptive traits within the population increases, and the frequency of non-adaptive traits decreases. The increase in the frequency of adaptive traits (a macro-evolutionary change) is merely a fortuitous consequence (though a predictable one) of the effects of the environment on the procreative success of individual organisms. The same holds for even larger scale evolutionary changes such as speciation. Historical materialism, on the other hand, does not explain changes in the mode of production (e.g. the change from feudalist to capitalist relations of production) as merely the fortuitous consequence of many isolated changes in the relations of production in which individuals stand. Rather, the fact that the relations in which individuals stand change is explained by the fact that the mode of production itself changed, and changes in the mode of production are in turn explained by the fact that the old relations of production had become fetters on the development of the productive forces (and, of course, the Development Thesis). Explanitorily speaking, large scale changes are prior to small scale changes in historical materialism.
Now the explanatory priority of large scale changes within historical materialism is one aspect of the theory that makes it suspect in a way that evolutionary theory is not. For one might object, couldn't some small scale changes in the relations of production occur without the mode of production undergoing a thorough change. Now of course in a purely logical sense, this is entirely possible, but the mere logical possibility of such changes does not itself undermine historical materialism, since it is a theory about actual historical development and change. But still, we might think there are actual historical examples of such changes, including perhaps the growth of the French burghers during late feudalism (an example that Marx himself discussed).
This is a challenge that historical materialism may not be able to meet. But it is important to recognize that if we think that it is not possible to explain small scale historical changes in terms of large scale changes, then we are necessarily led to the conclusion that there can be no systematic theory of historical development at all. We cannot construct a theory of history that systematically explains large scale changes as the results of small scale changes, because there is no mechanism analogous to natural selection to explain small scale historical changes. And it seems to me that it is even more unlikely that there is a mechanism of this sort than that some theory with the explanatory structure (though perhaps not the content) of historical materialism is true. So, if we reject the explanatory structure of historical materialism, then we are left with the view that, as Levine and Sober put it, "history is irreducibly heterogeneous...and therefore ineluctably atheoretical." This is the view that most people accept, and perhaps it is right. If so, then history simply isn't a proper subject for scientific study, and there can be no systematic theory of history. But as Levine and Sober rightly point out, "whether history admits of a general theory should be discovered, not settled by speculative fiat." Historical materialism may very well be false; but if it is, it cannot be proven so merely by citing the apparent heterogeneity of explanations of individual historical events.
posted by Brian Berkey at 4:00 PM
http://www.blogger.com/comment.g?blogID=18266847&postID=3573616402664919544&isPopup=true Comments:
What enormous categorical confusion, Brian.
"Marx's theory can be understood to have roughly the same explanatory structure." Evolutionary theory has no "structure." It consists of five separate theories, and if five interrelated theories constitute a "structure," your polysemy is showing.
Then you have the chutzpah to deny Marx's theory is teleological, and cite:
"according to historical materialism, selected for because they move history toward some end that it is in any sense "aiming at" (though this is an understandable misinterpretation of Marx given the fact that he refers to the transition to communist relations of production as marking "the end of history")." No misunderstanding. Every Marxist takes dialectical historicism precisely this way. You seem to have substituted historical materialism (where's this from?) in your confusion. Dialectical Historicism is entirely teleological. I have no idea what "historical materialism" is presumed to mean. But Marx was tad confused and confusing, structurally and functionally.
Finally, "science" is used equivocably. "Social science" is incapable of predictions, incapable of experimentation with controls, incapable of any "science's" conclusions, and thus is not scientific, not even economics is probablistic or "structured" (Laura Tyson and Brad DeLong will confirm this over at Haas, as will any Federal Reserve officer). Ergo, "structures" is a meaningless economic, historical, and even materialistic concept actually used to compare with a "scientific theory." What "structures of science" (itself entirely incoherent) cohere with Marxist "structures" is long past acknowledged to be FALSE. Marx's economic theories (which are not scientific), his dialectical materialism (which is biologically false), and his dialectical historicism - is also false, and whether those falsehoods lead to class warfare and the Communist Utopia no Marxist worth any salt (if any still exist) can refute or claim, because the CLAIMS themselves are false. (Please feel free to consult former Marxist Wayne Dynes and Leszek Kolakowski Main Currents of Marxism, wherein he writes of Marxism as the "greatest fantasy of our century." He taught at Berkeley before Oxford, and without appealing to authority, his 2004 update might be worth a consultation.) Let's just bury Marx (with Freud).
The Structure of Scientific Revolutions left Marx in a vortex of nowhere with Lysenko starving 20 million Soviets, Mao annihilating 20 million Chinese, not counting gulags and purges, so the Communist economics failed, as did the methods of production. BIG TIME.
Revisit Boyle's scientific method, modified by Popper, against his protege Feyerabend's objections, and then tell us why String Theory was abandoned because it failed the Popperian Falsifiability Criterion? Science has a "method," with "theories," based on "observation," that can, in principle, be falsified, so in practice they can be verified. When a theory describes the fact sufficiently, repeatedly experiments and data confirm the theory, we call it a "law." Boyle or Bacon can assist from the 15-17th C.
Gilbert Ryle had a phrase for this post: Category Mistake(s). Austin: Linguistic abuse. I don't need to tell you of Searle's.
The skeptical stance requires one question one's premises, before apply the rules of logic, because, if the premises are untrue, the conclusion that follows may be logically valid, but not logically sound. Whatever Marx and Freud tried to do met neither of them came close to meeting these standards, which might explain why Marxism is rarely heralded in Anglo-American philosophy.
GS,
First, when I refer to evolutionary theory's "explanatory structure," all I mean is that there is a particular way in which evolutionary theory explains evolutionary changes (e.g. large scale changes are the fortuitous result of many small scale changes, which are themselves explained by natural selection). To deny that evolutionary theory has a structure in this sense is absurd.
Second, you claim that every Marxist takes "dialectical historicism" to be teleological. This is at best a gross overgeneralization. Furthermore, dialectics is a METHOD of analysis, and one can accept the method without accepting any of the content of Marx's theories; this should be clear from the fact that Hegel was a dialectical thinker as well. Of course Hegel's view of history was teleological, but nothing in the dialectical method itself requires that one accept a teleological view, and even if Marx himself accepted such a view (which I doubt), it is surely possible to construct a theory that includes many of Marx's central claims but dispenses with teleology (see e.g. Cohen).
So since dialectics is a method and therefore does not necessarily involve teleology, perhaps it is Marx's historicism that you think is necessarily teleological. Popper seems to have thought that accepting a general theory of history required accepting a teleological view, but this seems wrong, since Cohen and others have developed general theories of history that are not teleological. We can distinguish, then, Strong Historicism from Weak Historicism. Strong Historicist theories are telelogical theories of historical development, while Weak Historicist theories are non-teleological general theories of historical development.
Now as Levine and Sober rightly point out, historical theories are not limited to those that attempt to explain the development of human society. Historical theories include any theories that attempt to explain development over time, and that include a clear direction in which that development is expected to move. And in this sense evolutionary theory is a historical theory, and a Weak Historicist theory. Theories that are directional in this sense need not be teleological; as Levine and Sober point out, evolutionary theory predicts that over time populations will be become better adapted to their environments as a result of natural selection, but this does not make evolutionary theory teleological. In the same way, a historical materialism that assumes Cohen's Development Thesis can avoid teleology, and so Cohen's view turns out to be a Weak Historicist view as well.
Granted, "dialectics" is as ancient as antiquity. No one denies that. Aristotlean and Hegelian dialectics are not synonymous, nor are Boole's. Modern mathematical logic is part of dialectics; whether inductive inferences are is a matter of dispute in the philosophy of logic. Because of Marx's contamination of the Greek term, most prefer "logic" to "dialectic," or even "heuristics," since Marxism has so many adverse and negative associations of use.
"Dialectics" and "dialectical materialism" are not synonymous, but you seem to suggest they are.
Your use of "historical materialism" must be from Stalin, not Marx or Engels. Maybe other Marxists use the phrase. Are you suggesting that Stalin's "historical materialism" has a structure akin to Darwin's five theories of evolution. If so, you should illustrate the correspondence, since it is not commonly known. Certainly, the Lysenko Debacle did not correspond to Darwin, but to Marx, starving 20 million in the process.
A Theory of Dialectic of Nature was altered in 1848, but as Kolakowsksi writes, "historical determinism, the theory of classes, of the state, or of revolution -- were understood differently be different people. This is the natural fate of all social theories without exception," and "no modern theory can rival Marxism for mutually inconsistent interpretations" (op. cit, 191).
Apparently, Cohn, Levine, Sober, et alia, are continuing in the inconsistent interpretations and linguistic use of Marxism, which, for me, at least, makes it too ambiguous to be useful. (Indeed, ambiguity only serves to perpetuate the casuistry and inconsistency obfuscation.) Perhaps this ambiguity "fits" Derrida's Rhetoric of Indeterminacy and Structuralism's binary oppositions without stable signification, that begets what is currently known as postmodernism. But at this point, it would be a literary and philological enterprise, not a philosophically interesting one. In fact, I'm not even sure what the issues and structures apply to.
But, notwithstanding the above, I still do not understand your claim: "evolutionary theory's "explanatory structure," all I mean is that there is a particular way in which evolutionary theory explains evolutionarly changes." Structures are an archetonic, while changes are a process. These are categorically quite distinct. Perhaps you could illustrate what you mean by "[a] evolutionary's theory's [b] explanatory structure = explanation [c] evolutionary changes."
Evolutionary theories (plural=five) do describe/explain various processes (again, plural), when regressively applied, explain the evolution processes of life, the origin of species, the random variations of mutations, the descent of man, by adaptationism and fitness. How is this statement "structural" in your intended meaning of apparent synonymy with "theory"?
GS,
Nowhere do I suggest that 'dialectics' and 'dialectical materialism' are coextensive; Hegel was an idealist rather than a materialist, and he employed the dialectical method, as I pointed out earlier. Also, "historical materialism" is the standard term used to refer to Marx's theory of history. It appears in the titles of a great many articles by analytic philosophers on Marx, as a JSTOR search for "historical materialism" will reveal. It is certainly not a peculiarly "Stalinist" term (in fact I believe it was used by Lenin long before Stalin), and pointing out that Stalin (who was without a doubt a horrifically evil man) killed 20 million people is not in any way an argument against taking Marx's theory of history seriously.
You're right that there are a great many mutually inconsistent interpretations of Marx out there, just as there are of Hume, Rousseau, Plato, and many other philosophers, all of whom still deserve to be read by contemporary students of philosophy; indeed often it is the conflicts over interpretation that make studying a particular philosopher most interesting. Incidentally, I have no idea why you think any of this relates to Derrida or postmodernism, and I have no interest in it - Derrida's work (and most postmodernist work) is incomprehensible nonsense, and Searle was right when he said that it "is the kind of stuff that gives bullshit a bad name."
On the point about explanatory structures, I'm merely pointing out that there are different kinds of explanations, and that they work differently. For example, causal explanations have a unique structure, and different causal explanations employ the same structure, filling it in with different content. And the same goes for other kinds of explanations (e.g. functional, intentional, etc.). In the post I was merely attempting to point out some of the similarities and differences between the explanatory structures of evolutionary theory and historical materialism.
Oh wow, I just found this! Will read carefully asap :-)
Brian,
Thanks for addressing my previous comment. At the end of the day, I agree that evolutionary theory is falsifiable. However, I remain interested in objections claiming that it is not falsifiable - not because I believe they are correct, but because I wish to defend evolutionary theory against such attacks. Critics seem to take evolutionary theory's vagueness of implications (which you mentioned) to be evidence of its non-falsifiability, and I fear that talk of "missing links" and fossils yet to be found does little to dispel the criticism. I wish there were a closer-to-irrefutable sense in which evolutionary theory could be understood as clearly falsifiable.
You are absolutely correct that it is traits which are selected for and not so much organisms themselves. That was poor and thoughtless word choice on my part.
I found it illuminating to consider evolutionary theory and historical materialism side by side. Without the former as a guide, the latter would have been wholly lost on me.
pjs
I would just like to point out that Popper's relation to Marx is more complicated than usually assumed. Popper was in fact a great admirer of much of Marx's own analyses. Bill Groton has written well on some of Popper's relationship to the intellectual tradition of Marx in his paper "Popper's Debt to Marx" which can be found here:
http://kiosk.polisci.umn.edu/information/theory/kiosk/spring2001/Popper.pdf
You might be surprised that Popper considered Marx a committed and honest scientist, but "Marxism" as a travesty.
Jerry
Jerry,
You're right that there is much in Marx's work that Popper admired, although he also thought that there were serious problems with his methodology that trace back to Hegel's influence on Marx. Thanks for the reference to the Groton paper; I'll have to take a look at it.
Post a Comment
Links to this post:
Create a Link
<< Home