Lit-crit and the scientific method

by Henry Farrell on June 25, 2009

Following Michael’s pointer, I read William Deresiewicz’s “piece”: with some interest – while I’m as happy as the next person to read good take-downs of dodgy ev-psych arguments, I found some of the claims a little … sweeping. Take, for example, the suggestion that:

Having colonized the social sciences–where it has begun to displace the view, predominant throughout the twentieth century, that the mind is a highly malleable product of culture–[Darwinian evolutionary thinking] has now set its sights on the humanities, the last area of resistance.

I’m sure that ev-psych types would _like_ this to be true1, but as a card-carrying social scientist, I have yet to be informed of the successful colonization of sociology, political science, economics and anthropology by explanations based on Darwinian theory. Nor, for that matter, did I know that economists _ever_ believed the mind to be a highly malleable product of culture.

But more to the point, Deresiewicz’s account seems to me to be a little bit intellectually loose. It vacillates between criticizing evolutionary psychology because its theories are scientifically poor, and criticizing it because we shouldn’t be trying to apply scientific methods to the humanities in the first place. On the one hand when Deresiewicz criticizes Boyd (whose work he clearly likes a lot better than the more ideologically opportunistic stuff available from Dutton etc), he argues:

the attractiveness of a theory is no brief for its validity. Because storytelling, absent literacy, leaves no record, Boyd’s reasoning rests entirely on analogy and deduction. Primates do this, children do that, contemporary hunter-gatherers do the other; therefore this is what primitive humans must have done. Fiction serves these functions now; therefore it always has. This kind of thinking may be clever, but it isn’t science. It also overlooks the crucial phenomenon of functional shift. What evolved for one purpose can end up developing many others.

I haven’t read Boyd, but this seems like plausible, useful and specific criticism – if Deresiewicz is right, Boyd has written a book with an interesting, if overly sweeping argument, but doesn’t have the evidence to back it up and to really make it scientific. So far, so good. But how does this sit with Deresiewicz’ later claim that we shouldn’t be talking about literature in these kinds of ways at all?

There are several things to say about this and similar work. For one thing, a lot of it isn’t literary at all. It simply uses literature as a source of data for social-scientific investigation, and it takes no cognizance of literary form–that is, of what makes literature different from other modes of discourse. For another, unlike fields such as criminology or public health that were transformed by the introduction of statistical methods, literary studies is not concerned with large classes of phenomena of which individual cases are merely interchangeable and aggregable examples. It is concerned, precisely, with individual cases, and very few of them at that: the rare works of value that stand out from the heap of dross produced in every age.


That so many of the greatest works of literary art–the Iliad, the Aeneid, the Divine Comedy, Don Quixote, Hamlet, King Lear, Paradise Lost, Faust, Moby-Dick, the novels of Dostoyevsky, Joyce, Woolf and Coetzee–are ultimately concerned not with mate selection or status competition, however seriously they might consider such matters, but with the human place in the cosmos; that such a commitment is precisely what begins to distinguish these works from the kinds of things that are better studied with polling data and cheek swabs; that the finest books demand a criticism that attends to what makes them unique, not what makes them typical: these are not possibilities that literary Darwinism envisions.

This is obviously a much more sweeping claim – it suggests that even if Boyd, and people like Boyd are right, they are wrong – that one simply cannot usefully understand what literature does unless you concentrate on the great books, one at a time, treating each and every one of them as providing unique insight into the human condition.

Now I don’t want to argue with the proposition that it _can be very helpful_ for certain purposes to treat each book as an unique, precious, delicate snowflake. However, it is rather obviously true that we can also arrive at very useful knowledge by thinking about how books and authors are situated in wider contexts, for authors and their books are indeed situated in a variety of ways – their times, their themes, their engagement with other books in similar or dissimilar genres. This situatedness is often crucial to a proper understanding of the books. And not only are books situated, but they (including great books, however you want to define them), can very often be _classified_ in ways that are useful both (a) to the understanding of the book in itself, and (b) to the understanding of the book as something situated in a social space of dialogue or whatever you might prefer to call it. No man is an island, and no book is, either.

I’m not even being controversial here – lots of work out there in literary studies that does this kind of work without considering itself to be at all unusual or odd. Just yesterday, Miriam Burstein mentioned in passing that it is useful to think of “Daniel Deronda”: in the terms of the nineteenth century genre of Jewish conversion fiction, the key tropes of which it (deliberately?) reverses. And if you accept that it is sometimes useful to study Daniel Deronda‘s relationship to a wider genre of fiction that was popular at the time, then you are, I think, accepting that it is sometimes useful to study works of fiction as members of ‘large classes of phenomena’ or as being related to these phenomena in significant ways.

And if you accept this – why then, as John Holbo suggests in his “friendly rejoinder”: to Jenny Davidson, you are halfway to conceding that quantitative methodologies a la Franco Moretti can actually be helpful to the understanding of literature, as well as more qualitative forms of soaking and poking. And this (to make clear the destination to which this slippery slope is propelling us) suggests, in a backhanded kind of way, that we may be back to the claim that evolutionary theory, of a kind, is quite useful to the understanding of literature. I give you “Cosma Shalizi:”:

The small-scale details of literature and of human life have an intrinsic interest and value that is missing from the small-scale detail of molecular chaos, so there is certainly all the room in the world for what Moretti would like to do and close reading, and even essayistic appreciation. (But there is not, I am afraid, room enough in the world for Harold Bloom.) …

When trying to explain cultural change and cultural variation, people have generally sought to do so by supposing culture is causally driven by something else (the climate, the social structure), or, even more strongly, that it is adapted to something else, or, more strongly yet, that it functions adaptively for the benefit of something else … This has led to an awful lot of (if I may use the phrase) adaptationist just-so stories, and uncritical analogy-mongering on a level with the sort of thinking which leads rhinoceros horn to be prescribed for impotence. …

Checking hypotheses about causation, and still more about adaptation, is really hard with just one case, arguably hopeless. What you need is the ability to reliably detect departures from the hypothesis, if they are actually present — “power”, in the statisticians’ jargon. It is hard to get much power when n=1. If you want to claim that certain aspects of 19th century British novels were the way they were because those features fitted with ideologies of British imperialism — a fairly strong hypothesis about adaptation — I don’t see how you can do it just by interpreting Mansfield Park, no matter how subtle and sophisticated your reading. On the other hand, if you look at lots of contemporary novels, and the ones which (say) depict Great Britain’s relations with its colonies in the same way as Mansfield Park does are systematically more successful, on average, than those which depict it differently, well then I don’t see how that couldn’t be good news for your idea, though even that would really only be the beginning of backing it up.

Biologists have given a lot of thought to checking hypotheses about adaptation, and developed many means of doing so. Mutatis mutandis, many of these means could also be applied to literature, or other aspects of culture. …

The foregoing shouldn’t be taken to mean that comparative literature should slavishly imitate comparative biology. There are people who have thought about the application of evolutionary ideas to social and cultural change in ways which are much more sophisticated about psychology, social organization and human interaction than (most) advocates of memetics; I am thinking particularly of David Hull, W. G. Runciman, Dan Sperber, Stephen Toulmin’s great The Collective Use and Evolution of Concepts, and even the fragmentary MS. of Adam Westoby. As the economist Richard Nelson writes, we should expect our ideas of general evolution to change as we learn more about cultural evolution. We should also expect to have to develop different methods of data analysis. But, as always, we start with what we already know how to do. …

I don’t see the harm in trying to make this all fit together as another instance of a general pattern, alongside biological evolution, because they have similar causally-relevant features, and so similar mechanisms are at work. Many people have pointed out, in some detail, that explaining biological processes through the joint action of variation and selective transmission in populations is one instance of a general pattern of historical explanation; Toulmin is particularly clear on this. There is a demography of businesses, of interest groups, even of medieval manuscripts of classical works, and so why not one of literary texts? Inheriting discrete, particulate hereditary factors from a small, fixed number of immediate ancestors is not the sine qua non of this form of historical explanation, though the details of the process of inheritance will very strongly affect the character of the resulting dynamics. It might be that theories of literary change cast in this form are too complicated to be useful, or that we just don’t know enough yet to find the useful ways to formulate them. But it wouldn’t hurt to seriously try, and we’d learn a lot, no matter the eventual outcome.

So, as I interpret what Cosma is saying here, a good evolutionary account of literature would very likely tell us interesting things that we don’t know already. However, such a theory would very likely be different from standard evolutionary psychology accounts – the mechanisms of selection and retention underlying cultural change are likely to be very different than those that underlie evolutionary biology. Nor is the one easily reducible to the other – even if we are evolved beings, many of our cultural forms likely serve no useful biological function.

This also implies a specific subsidiary point – that Deresiewicz’ ‘we need to forget about scientific methods when talking about culture, and instead look at how great authors have unique insight into the world’ argument can have specifically pernicious consequences for public debate. I don’t have any particular objection to loose sociological and cultural argument that doesn’t appeal to evidence, testable hypotheses etc. Often, it is going to be the best that we can do – even in the unlikely event that every culturalist/literary theorist converted to the program that Cosma proposes above, there would be swathes of important social and cultural questions that we wouldn’t have the evidence or methods to provide sound answers to. But I _do_ object to forms of cultural criticism which ground their claims in the argument that the great authors know best, suggesting that we _don’t need_ to look at the world to understand it; all we need to look at is what great authors say about the world. Deresiewicz would plausibly say that he isn’t advancing this claim; all that he is suggesting that great books tell us about man’s place in the cosmos, not men and women’s place in society. But unless we limit literary criticism to the explication of a very, _very_ narrow conception of ‘man’s place in the cosmos,’ speculation about these universal truths is likely to come encumbered with a whole lot of sociological baggage. The magisteria of social science and literary criticism overlap with each other, and did before the advent of cultural studies (it is hard to imagine, given their major interests, how they could not while remaining interesting).

For an example of just how badly this can go wrong, see this “recent piece”: by Adam Kirsch (who is far from the worst of the conservative literary critics) on Europe’s purported lack of the manly virtues celebrated by timeless philosophers such as Plato, Nietzsche and Francis Fukuyama:

Is it true that Western Europeans, after half a century of peace and prosperity, suffer from the kind of moral malaise that Nietzsche warned about, and that Fukuyama and Kagan diagnosed? One way to answer this question is to listen, not to American pundits, but to Europeans themselves—in particular, to their novelists. In the nineteenth century, a reader of Dostoevsky and Flaubert could have gained insights into the state of Europe that a reader of newspapers would have missed. In the twenty-first, it is at least possible that the most significant European novelists can give us similar insights. … when novelists from different European countries, writing in different languages and very different styles, all seem to corroborate one another’s intuitions, it is at least fair to wonder whether a real cultural shift is under way. …

The three novels I wish to consider are not, of course, anything like a representative sample of the fiction being written in Europe over the last two decades. But W. G. Sebald’s _The Rings of Saturn,_ Michel Houellebecq’s _The Elementary Particles,_ and Ian McEwan’s _Saturday_ are as distinguished and emblematic a selection as might be made. All of these writers were born in the 1940s and 1950s, and emerged as major novelists in the 1990s. In other words, they are members of the post–World War II generation, and did or are doing their most important work in the post–Cold War period. They belong to, and write about, a cosmopolitan, peaceful, unified Western Europe … they in no sense form a school or movement. … Three more different writers could hardly be invented. Which makes it all the more suggestive, I think, that their portraits of the spiritual state of contemporary Europe are so powerfully complementary. …

But just as Shakespeare tells us that a sad tale’s best for winter, it is precisely the wishfulness of _Saturday_ that makes it an apt parable for Europe’s own Saturday—the last day of the week, the day of rest. This metaphor holds equally well for Houellebecq’s novel, with its nightmare of perpetual pleasure-seeking, and for Sebald’s, with its reverie of retrospection: both seem to take place in a civilization that has retired from its historical tasks, having done and suffered so much that further effort seems impossible. Yet actual history, of course, does not allow for days of rest, and the historical world of which Europe is a part will not release it into the gentle euthanasia of Houellebecq’s imagination, or the quiet senescence of Sebald’s. McEwan’s vision of a civilization dragged back into conflict and struggle, by foes more brutal and irrational than itself, seems much more likely to resemble the actual future. That is why McEwan’s inability to imagine a realistic victory in that struggle makes _Saturday_ perhaps even more troubling

You can see in these passages some dim inklings of an understanding of the scientific method – as Kirsch acknowledges, the authors picked are not a random sample. But his proposed focus on authors who are ‘distinguished’ and ’emblematic’ while not belonging to the same ‘school or movement’ is less an alternative to that method than a rhetorical flourish. There is no reason to believe that Houellebecq has any more insight into the state of European politics and society than did W.B. Yeats – and while time may have pardoned the latter’s political views, it hasn’t made them any more right than they were at the time. That Ian McEwan apparently agrees with Houllebecq (Sebald’s _The Rings of Saturn_, which is a book I know and love, doesn’t really belong in this review at all) perhaps says less about the state of Europe than about the more “specific historical convergence”: between a small group of prominent English writers who want a muscular anti-Islamic rationalism and their new-found traditional xenophobic conservative allies.

Because Kirsch grounds his argument in the authority of universal truths of the human condition as revealed to giants of modern literature, he never has to deal with inconvenient facts that might complicate his analysis. Kirsch doesn’t ground his account of the unmanly weakness of modern Europe in any actual discussion of European society (unless you count his brief nods in the direction of Kagan etc). He clearly feels that he doesn’t need to do any more than to read a couple of ideologically congenial novels, proclaim them to have profound insights, and hence derive a general spiritual diagnosis of the Old Continent’s desuetude. The problems of this approach (selection bias, strained readings etc) are too obvious to detail at any length and are immediately apparent to anyone with even a cursory acquaintance with European intellectual debates, let alone changes in European societies.

I have no particular reason to believe that Deresiewicz himself would approve of Kirsch’s method here (I don’t know Deresiewicz’s work at all – not my field). But if (as Deresiewicz seems to me to be quite explicitly suggesting) the scientific method has _no place_ in literary criticism because it cannot encompass the great and unique insights of a few key authors, then portentous-yet-entirely-ungrounded pronouncements a la Kirsch are unavoidable. Genius (as Deresiewicz himself explicitly recognizes in the earlier part of his essay) is no substitute for hypothesis testing. Yet a ‘unique insights of great novels’ approach carries the implication that genius can and should be recognized as a specific source of authority – and it is difficult for me, at least, to see, how the kind of authority that Deresiewicz invokes can easily be separated from the kind of authority that Kirsch is trying to invoke here. Kirsch surely believes himself to be involved in the explication of timeless truths, expounded by writers of genius – that they can be readily applied to Europe’s current condition, as diagnosed by distinguished contemporary novelists, is merely further proof of their general applicability. And the point here (at least for me) isn’t that Kirsch is wrong (although I believe he is). It is that this kind of invocation of authority is no more than a rhetorical trick, a spurious intellectual conceit. I suspect that the Deresiewicz of the earlier part of this essay would agree, even if the Deresiewicz of the later part would not. And therein lies the major internal contradiction of this essay.

1Deresiewicz’s description of what is at stake in this battle sounds suspiciously similar to Cosmides and Tooby’s guff about the war between the forces of truth, light, and evolutionary psychology on the one hand and the cultural determinism of the so-called ‘Standard Social Science Model’ on the other. The latter being, as I have noted before, neither standard, nor social scientific, nor for that matter, a model.



Donald A. Coffin 06.25.09 at 3:55 pm

Henry writes: “Nor, for that matter, did I know that economists ever believed the mind to be a highly malleable product of culture.”

Well.. yeah, actually we do. We assume that people’s preferences affect their choices, but we have no theory of preferences. So we make no assumptions about them at all…except that they are malleable and probably affected by the culture in which an individual lives. (Or at least that was the state-pf-play whenI was in grad school.)


Henry 06.25.09 at 4:02 pm

Donald – the ‘de gustibus’ assumption does not equate to malleability, at least in the form that Cosmides, Tooby etc treat it. Preferences are not only exogenous – they are fixed. Furthermore, economists tend to assume (although they don’t universally do this) ‘narrow’ rationality – that is, that actors’ preferences have a lot to do with whether they would be economically better off or worse off given different outcomes.


Donald A. Coffin 06.25.09 at 4:18 pm

Not wanting to belabor this, but I disagree that economists think preferences are fixed. We just think they change in ways that we’re not competent to explain/explore, so we generally rule changes in preferences out-of-bounds as an explanatory option. (Again, we have no theory of preference formation.) And we do that because changes in preferences are like a massive “get out of jail free” card–you can explain almost any behavior by saying “Well, preferences changed.” So the disciplinary standard is to assume unchanged preferences and look for other explanations–but be willing to accept the hypothesis of changed preferences if there are no other explanations. (And, speaking for myself, if psychologists or sociologists have evidence of changed preferences, I’m happy to accept that…and it is, after all, one of the things about the behavioral economics movement that makes it different.) (Again, the stuff about disciplinary norms is as of when I was in grad school. Which is longer age than I like to think about.)


Salient 06.25.09 at 5:06 pm

Nor is the one easily reducible to the other – even if we are evolved beings, many of our cultural forms likely serve no useful biological function.

Neither do many of our biological forms (e.g. the design of our circulatory system would flunk a first-year mechanics-and-fluid-dynamics student, as anyone who has attempted to stand up very quickly is aware).

Apologies for picking the nits on a great post.


LFC 06.25.09 at 6:16 pm

Although Kagan’s argument that “the U.S. is from Mars, Europe from Venus” is, I think, probably at least two-thirds rubbish, and while I therefore have little sympathy for the substantive conclusions that Kirsch comes to, it seems to me that Kirsch’s method, as opposed to his specific arguments, is not as objectionable as your post suggests. Based on reading the excerpts from Kirsch that you quote and glancing at the full article, he appears to be doing something literary/cultural/social critics have been doing for a long time. He reads a few novels by well-known writers and proceeds to consider what they might imply — on his reading — about the condition of contemporary society. His language is cautious — “one way to answer this question” (emphasis added), “one begins to wonder whether this might not [etc.]” — suggesting that he knows he is not scientifically demonstrating anything (or, AFAIKT, expounding universal truths) — but just doing a fairly standard “here is what some novels, in my opinion, appear to tell us about X, take it for what it’s worth” kind of piece. I would be surprised, as I say, if literary critics had not been doing this for a very long time, and to suggest that this procedure amounts to invoking the authority of timeless genius to ‘prove’ a sociological proposition seems to me to involve some kind of category mistake on your part. Those who read, say, Dickens or George Eliot or Zola for their insights about the society in which they lived would not claim, I think, that b/c these authors are geniuses you can find out everything you need to know about nineteenth-century debtors’ prisons just by reading Little Dorrit or everything about 19th c. French coal mining just by reading Germinal. Similarly, I would think Kirsch would concede that the validity of Kagan’s thesis cannot be definitively established just by reading Saturday and the other novels he considers. But that doesn’t make his “method” illegitimate.


Henry 06.25.09 at 6:52 pm


(1) where do I say or even imply that literary critics have not been trying to do this for a very long time? I think that my piece rather implies the contrary. And where on earth is the category mistake. As I note, the borders between literary criticism and social speculations are at best poorly defined. That this has a long history does not, however, justify it.

(2) It is quite clear from the piece that despite his qualifications (which I include in the bits I quote), Kirsch _is_ making a general spiritual diagnosis of the European malaise – see in particular the discussion at the end about the likelihood that Europe will come under attack by foes more brutal and irrational than itself (I don’t think one needs much imagination to figure out who these foes are). I simply don’t think that you can credibly portray this as a ‘take this for what it is worth’ piece – its tone throughout is rather ‘I can’t prove this satisfactorily, but I know it to be true.’ When you use words like ‘suggestive’ you are not saying ‘maybe’ – you are intimating that you are talking about evidence that likely tells us something about the true state of the world. And the point is that either which way, the proffered evidence is _useless_ – it involves the selection, on the part of Kirsch of two ideologically congenial authors, and a third (Sebald) whose work can be made to sound as if it fits with the first two with a little bit of inventive squeezing. Obviously you could make radically different arguments about Europe’s current state by picking other authors. That Houllebecq and McEwan have this or that _opinion_ do not demonstrate that they have any specific _insights_, much less ones that can be generalized. To put it another way – what do _you_ think that Houllebecq and McEwan’s opinions on the current state of Europe have to tell us about what is actually going on? What kinds of useful evidence do they present of any phenomenon beyond the likely fact that Houllebecq and McEwan (and, at a pinch, writers and people like them) believe (at least in part because of their specific and contingent circumstances) that they live in a lily-livered society threatened by the thugs from beyond? I don’t believe that we should give Houllebecq or McEwan’s opinions any evidentiary status beyond that of the opinions of other moderately well informed and reasonably articulate observers. Kirsch clearly disagrees – otherwise he would not have written this terrible article. And I believe that Deresiewicz’ stance gives a considerable degree of license to people like Kirsch to make these kinds of claims.

(3) On universal truths, importance to Kirsch thereof, see the discussion of the specifically manly virtue of _thymos._


Steve LaBonne 06.25.09 at 7:06 pm

I’ve always felt that the only useful and defensible attitude toward any set of tools in any field of scholarly inquiry is eclecticism. You want to use whatever tools are available that seem to have promise of yielding interesting results, and at the same time you never want to confuse the particular toolset you’re using at a particular time with the inquiry itself. So as a complete outsider to literary studies I would go out on a limb and hazard that the most sensible answer to “what is the place of Darwinian arguments in literary criticism” is unlikely to be either “they’re always and everywhere totally useless” or “they’re the only way to go” but more likely “they might be of some value in some circumstances, give it a go and see what you come up with”.

Now having said that you want to use tools that are themselves solidly validated (it’s hard to pound nails with a hammer that has a broken handle), and in that regard there certainly are a lot of open questions about evo psych.


LFC 06.25.09 at 7:32 pm

I will defer a real response (if any) until I’ve read the Kirsch piece properly. Whether novelists’ opinions (insights?) should have any “evidentiary status beyond that of the opinions of other moderately well informed and reasonably articulate observers” is a question we might still end up disagreeing on, but I’ll have to think about it.
That literary critics have been doing something for a long time does not, of course, make it right or justifiable. But part of the problem may be that it is more hazardous to rely on contemporary writers to support arguments about contemporary society, as opposed to drawing on long-dead authors to buttress (or give added depth to) historical claims. In other words, we don’t yet know whether McEwan, Houllebecq, Sebald or whoever will seem, in retrospect, to have had “insights” as opposed to “opinions.” Whereas for eras that are long gone, (perhaps) it’s clearer. I’m just speculating here, obviously.


Henry 06.25.09 at 7:48 pm

LFC – I do think that relying on long-dead writers for _post-hoc_ insight is more defensible (in part because we can see how some of those insights stood the test of history). I also think that writers like Zola in particular can be helpful in understanding their epoch (because of their relentless interest in the quotidian stuff which doesn’t make it into the history books). So, perhaps we agree here. But of course, many of these insights are quite different from the ‘what should we do about Europe and Islam’ insights that I think Kirsch is trying to extract from these writers.


John Quiggin 06.25.09 at 8:46 pm

On economics, the standard rationality assumption is stylistically very similar to that of Ev Psych (in particular, not malleable, and based on a hardnosed assertion of unrestricted optimisation) but substantively very different, since the core point of Ev Psych is that selfish genes promote kin altruism, which may then extend to group altruism.

I’ll point back to this old post observing that lots of people are attracted to the hardnosed style and simultaneously advocate “selfish gene” Ev Psych, “selfish person” neoclassical econ and “selfish state” international realism. But each of these explanations is totalising and can’t coexist with the others. If people are driven by their genes, they’ll be altruists at least some of the time, and if states are collections of self-interested individuals they won’t themselves act like self-interested individuals (see Arrow etc).


Henry 06.26.09 at 1:46 pm

What John says, basically (or at least, that is my understanding of econ’s underlying assumptions). And also what he says about the incompatibilities between the explanations.

Reading through my comments above though, it seems to me that I come off as being a bit too totalizing. I can happily accept that novelists can have interesting and original sociological insights – but these insights, like all other insights, need to be able to withstand the test of evidence. If Kirsch’s piece had started with the novelists, and then gone on to argue that this or that aspect of European society showed they were right, it would have been a better piece (although still a piece I would have had problems with, since I think it is, apart from its style of argument, based on some badly flawed ideas about what is happening in Europe).


Bill Benzon 06.26.09 at 3:40 pm

Yes, Deresiewicz opposes any intrusion of science into literary criticism. It’s a long-standing argument and is, IMNSHO, nonsense. You use what works and try to do better than the previous generation.

So, as I interpret what Cosma is saying here, a good evolutionary account of literature would very likely tell us interesting things that we don’t know already. However, such a theory would very likely be different from standard evolutionary psychology accounts – the mechanisms of selection and retention underlying cultural change are likely to be very different than those that underlie evolutionary biology. Nor is the one easily reducible to the other – even if we are evolved beings, many of our cultural forms likely serve no useful biological function.

The literary Darwinists seem to be, at best, indifferent to evolution in the cultural realm, though Boyd does have allusions to the possibility of “Darwin machines” (his term) operating within culture. Conceptualizing cultural evolution is proving to be difficult – memetics has gone nowhere – but that’s not a particularly good reason not to think about if it is culture that interests you.

Here are some remarks I made in a recent Valve post on Joseph Carroll’s ambition to completely revamp the study of literature on Darwinian lines:

I will note that I’m consistently puzzled about what’s evolutionary or Darwinian about this approach, whatever its intellectual merit may prove to be. It’s never been clear to me just what makes evolutionary psychology so, well, “evolutionary.” Oh, I know the rationalization, it’s about our evolved human nature. And surely we’re evolved, no? But, it’s one species, just one, and the evolutionary psychologists have (conceptually) stopped the clock on it. So there’s not much evolution in this psychology.

The application of those ideas to literature, however, changes things a bit. Now you’ve got a subject matter with a real history and lots of forms, lots of morphology. Here you could be sensibly evolutionary – though you’d have to conceptualize evolution in the cultural realm – and the Darwinists ignore the possibility. I see no interest in literary form, precious little in language, nor in literary history. And yet this is supposed to be the source of a revolution in the study of literature.

I don’t buy it. As far as I can tell “evolution” and “Darwin” function mostly as intellectual brand identifiers and little more. The actual thinking being done owes little to either. As for making the world anew, I don’t believe that is on the agenda, not the way I read the current practical criticism of the Darwinists.

Now let me offer an alternative view of the future of literary studies. I agree with Carroll that explicit steps need to be taken to assimilate the newer psychologies and I believe that perhaps in fifteen or twenty years literary studies will be conducted within a different framework. But I do not believe literary Darwinism will be the primary source of that framework. . . . I venture to say that evolutionary thinking will play an important role in that framework, but it will be cultural evolution, not biological. The fundamental terms of an approach to cultural evolution have yet to be worked out, though there is a good deal of work going on of various kinds (for my own highly biased view, look here) nor do I think they will have been worked out in twenty years. But we will be further along than we are now. Cultural evolution necessarily implies cultural difference, and so the last three or four decades of work on ethnicity, colonialism, and related matters will certainly have a home here, as will historicism. Nor can I see gender issues disappearing as they have a non-trivial cultural component.


lemuel pitkin 06.26.09 at 4:14 pm

lots of people are attracted to the hardnosed style and simultaneously advocate “selfish gene” Ev Psych, “selfish person” neoclassical econ and “selfish state” international realism. But each of these explanations is totalising and can’t coexist with the others. If people are driven by their genes, they’ll be altruists at least some of the time, and if states are collections of self-interested individuals they won’t themselves act like self-interested individuals (see Arrow etc).

This is a very good point, very well put.

But it does beg the question: so, then? To me, the obvious answer is that it is wrong to think of human beings or institutions as maximizing anything in general, tho obviously it can be a useful heuristic in certain contexts. But I’m not committed to a methodology which wants maximizing to be universal.

Is your view just that “selfish gene” and “selfish nation” are wrong but “selfish person” is right?


Chris 06.26.09 at 7:31 pm

lots of people are attracted to the hardnosed style and simultaneously advocate “selfish gene” Ev Psych, “selfish person” neoclassical econ and “selfish state” international realism. But each of these explanations is totalising and can’t coexist with the others.

How do you figure that? I’m not that familiar with selfish state, so I don’t know how states not named Louis are supposed to decide what they want (in order to selfishly pursue it in their relations with other states), but neoclassical econ pretty much doesn’t even attempt to understand where a person gets his preferences. If his selfish genes give him a preference for seeing his kin succeed, then he will attempt to satisfy this preference with behaviors like parenting and nepotism (which would seem difficult to explain without assuming some kin-oriented preferences).

When we get to states, I’m not so sure. It may be a category error to adopt the intentional stance toward a state, since (much more than a human or a dog) it is a collection of entities with discordant goals. (It would obviously be a category error to adopt the intentional stance toward a gene, but “selfish gene” theory doesn’t actually do so.) In any case, the gene-person relationship and the person-state relationship are very different.

It seems to me that these explanations (at least the first two) are no more discordant than chemistry is with physics; they explain different phenomena at different conceptual levels and in fact some of the answers at one level become inputs at the next.


John Quiggin 06.26.09 at 8:24 pm

LP, in a purely mathematical sense any pattern of behavior can be represented as maximizing something. See also. But I don’t support any of the simple models I’ve described.

Chris, you can restate the EP view in its strong form as “it’s a mistake to attribute the intentional stance to individuals, since they are the product of collections of entities (genes) selected by their capacity to reproduce themselves”. That, and not attribution of the intentional stance to genes, is the point of the “selfish gene” phrase.

Coming to neoclassical economics, there’s nothing to stop you doing neoclassical-style economics with objective functions incorporating altruism, or all sorts of other features inferred from EP. And, in the context of competitive markets, you can derive neoclassical-style results. But if you take preferences like this and apply them in other settings (for example, rational choice pol sci) you don’t get the standard answers.


nick s 06.26.09 at 8:36 pm

There’s a metanarrative of how lit-crit is taught, at least in the current Anglo-American tradition: you start with a basic education in narrative and character and prosody, get a dose of Richards/Leavis/NewCrit texts-in-themselves towards the end of your secondary education, and dive into the world of historical/genre/theory-based contextual criticism at university. That pedagogical progression reflects, to some extent, the archaeology of “English Literature” as a distinct field, and that field is certainly open to an analysis of its own.

I found Deresiewicz’s piece pretty banal, though: its opening swipe at theory is tired and predictable buzzword bingo, and as a result, his critique of literary Darwinism — juxtaposed with his “Theory” strawman — seems somewhat besides the point. Perhaps I’ve been out of the academy too long and I’m missing important distinctions, but his summary of Boyd’s hypothesis — “Fiction… is the way we train our minds for the vital business of social existence” — prompts me to wonder what’s new here other than the evo-psych framing. We tell stories to make sense of the world? No shit, Sherlock.

Steve LaBonne (with whom I’ve had a few disagreements here on different topics) sums up my basic attitude coming from the lit-crit side. Theoretical frameworks are toolkits, and they’re only useful to the extent that they yield interesting results, and those results needs to be judged in terms of the applicability of the analysis. My gut sense w/r/t “Darwinian” or “evo-psych” analysis in the abstract — again, I’ll profess my ignorance of the books under review — is that it’s basically putting a gloss on established social and cultural ones, though I’m willing to be convinced otherwise.


Tim Wilkinson 06.26.09 at 9:40 pm

JQ @15 – My Avast! antivirus software thinks your site (the first link: rationality-wednesday) is malicious, and blocks it. Obviously this could be a false positive e.g. based on some precautionary heuristic , but thought you might want to know. Though I’ve no idea what you would do about it…


Tim Wilkinson 06.26.09 at 9:49 pm

Tried to look at the Google cache for the last link (voting) and got a specific Trojan Warning:

File name:\{gzip}
Malware name: HTML:RedirBA-inf [Trj]
Malware type: Trojan Horse
VPS version: HTML:RedirBA-inf [Trj]

Again could be false positive but seems a bit more specific.


John Quiggin 06.27.09 at 10:35 am

I had a huge spam attack not long ago. It looks as if it has either returned or was never properly cleaned out. I’ll check on this.


magistra 06.27.09 at 11:21 am

The big problem with humanities types using Evolutionary Psychology as a tool is not that it’s dodgy science (as I said on an earlier thread, there’s a lot of dodgy ex-science, like Freud that stays in the humanities tool-box). It’s that EP is deeply sexist and been happily pounced upon by political reactionaries who want to explain why men and women are eternally different. I think a lot of academics in the humanities would be very unhappy with using EP given that aspect of it.

There was a discussion a few months ago on this site (see comments 70 onwards) about how a new form of ‘evolutionary psychology’ (without the capital letters) was replacing EP, and I think that is already generating stuff that will potentially interest people in the humanities (for example, experiments on differences between human and primate responses to music).


Bill Benzon 06.27.09 at 4:07 pm

David Sloan Wilson has a useful little piece in the Huffington Post that covers both the history and the current state of play in evolutionary psychology (of which Evolutionary Psychology is only one school).


John Quiggin 06.27.09 at 8:58 pm

Tim, if you have any more details, could you email them to me?


Tim Wilkinson 06.27.09 at 11:48 pm

That’s basically all I can find out from my cut-down freeware version of Avast. I can send you what log info I have (though that is really just what I’ve already posted) and a screenshot of the alerts raised if that might be of any help. If so, what is address (or rule for deriving it)?

It may be a false positive, since only the Google cache (from 21 Jun 03:20 GMT) raises a Trojan alert, and your site is merely described as ‘malicious’ – so maybe your site is still blacklisted even though the actual problem is fixed. I am in sync with the latest version of their database, and not at work or anything (i.e. there is no censorship software, of the kind I have encountered before, that blocks e.g. Al Jazeera).


Bloix 06.28.09 at 6:21 pm

“That so many of the greatest works of literary art … are ultimately concerned not with mate selection or status competition … but with the human place in the cosmos …”

This is an express claim that literature is a form of theology. If it means anything at all, it is a reference to the relationship between God and Man. (There is some literature that investigates the human place on Alpha Centuri, but most of it is not literary art.)

I personally don’t believe that most great authors are particularly concerned, consciously or unconsciously, with the human place in the cosmos. There are explicitly Catholic authors who have that concern, and some of them produced great works of art, but imho concerns with the human place in the cosmos tend to reduce the greatness of a work of art, not vice versa.

I think an author’s aspiration to explore the nature of a good marriage, or a bad one, or to investigate the constraints and possibilities allowed to an individual by the confines of the social order of the day, are quite sufficient goals for literature.


Tim Wilkinson 06.28.09 at 9:44 pm

the gene-person relationship and the person-state relationship are very different
but the gene-person relationship isn’t of course the only model for a person as less-than-perfectly-integrated, so that doesn’t settle the question of whether a state behaves like a person. If it does it’s probably a fairly pathological person, admittedly – or one whose ego-elite has effectively cut its id-masses out of the decision process?

Perhaps contra John Q, a state (qua international actor) is not a collection of all its citizens/subjects but only of the elite political, diplomatic and military personnel, who may be much more unified in their interests and approaches thus able to coordinate to form unified goals (interests), with only occasional ‘conflictedness’.

In any case, it’s one thing to say that the current theory of ‘selfish’ genes is incompatible with that of selfish people, another to say
the standard rationality assumption is…substantively very different, since the core point of Ev Psych is that selfish genes promote kin altruism, which may then extend to group altruism

Because the altruism in question happens ‘one level up’, so such a difference would have to be based on the further posit that groups of selfish persons can’t be (expected to be) altruistic to other groups (or groups of groups, depending on your reading of ‘group altruism’ as applied to the organism/person). Though there is currently no replication of persons across groups, so the ‘self’ in question is very different from that of a gene, which certainly makes a big and very relevant difference.

I’d also note that if group altruism is displayed by the gene-vector (organism) only as an unnecessary side-effect of kin altruism and not in itself because of a gene’s interest in replication, the gene is to that extent ‘irrational’. At this point the |selfishness| metaphor’s inutility is put into sharp relief – what is the point in positing selfishness of a being which doesn’t make any decisions?

And there’s a something else not really addressed here or in the other current RCT threads: the notion of ‘revealed preference’ which is a massive turd in the pool for economists. In effect it renders the notion of self-interest or indeed any interest contentless, and thereby does the same for rational choice. But luckily that fact is not really paid much heed by economists (Sen aside), who are happy to use it as a pretext for ignoring everything except (presumptively) selfish (better, non-tuistic) economic transactions when divining real, contentful interests. After all, people can’ be trusted to represent their preferences or non-preference-based motives in any other way (because thay are such selfish bastards, you see.) in lucro veritas! That is the motto of the von Neumanist-Morgansternites!

* Analogy between ‘preference’ as in ‘revealed’ and ‘fitness’ as in ‘survival of’ (though the latter needn’t feature in so-called selfish gene theory.
*Said theory only differs from a straightforwardly tidied-up conception of Darwin+genetics in its extraneous, and shall we say somewhat undermotivated, addition of a genetic determinist thesis about human behaviour.
*The behaviour of firms is another example of posited selfish composite-person behaviour, and with a similar legal/institutional underpinning to that which applies to states as international actors. (Is it OK/unambiguous to use ‘states’ in the US for non-domestic applications?)


Tim Wilkinson 06.28.09 at 10:28 pm

magistra @20: The big problem with humanities types using Evolutionary Psychology as a tool is not that it’s dodgy science (as I said on an earlier thread, there’s a lot of dodgy ex-science, like Freud that stays in the humanities tool-box). It’s that EP is deeply sexist and been happily pounced upon…

First point – context suggests you aren’t opposed to ‘dodgy ex-science in the humanities tool-box’. That’s a bit odd on the face of it, though there are no doubt subtleties.

What seems to be a similar approach has been canvassed @16 by nick s: Theoretical frameworks are toolkits, and they’re only useful to the extent that they yield interesting results, and those results needs to be judged in terms of the applicability of the analysis.

Steve LaBonne @7: “they might be of some value in some circumstances, give it a go and see what you come up with”…Now having said that you want to use tools that are themselves solidly validated (it’s hard to pound nails with a hammer that has a broken handle) gestures toward the unresolved tension here.

So are these unscientific tools merely heuristics or perspectives, and if so don’t they need to be eliminable from the argument? I.e., if in a given case you think they do apply under some interpretation, don’t you need to explain how and why? And if so, might they not just as well (or indeed prefereably, at least as a precaution against unclarity) be eliminated from the exposition? If a framework other other tool isn;t good enough that you can rely on it at least enough to defer its justification to a cited work, is it really any gooda t all? Or is it rather at best just a recipe for error and obfuscation?

2nd point – on the sexism of EvoPsych, surely the claim is that reality is in the same way (i.e. merely statistically, non-normatively and perhaps superably) sexist? After all it is with respect to upper body strength, say. I appreciate there may be abuses, but shouldn’t they be exposed as such rather than taking what might be regarded as an excessively precautionary approach or even imposition of a taboo?

Obviously there are subtleties here too – I suppose possible (more extreme?) comparators might be racial IQ comparisons and holocaust denial, the latter both properly-so-called, and in the extended ADLed-up sense that covers relatively minor instances of what could unloadly be called ‘revisionism’ if it weren’t for the fact that both holocaust deniers and holocaust-history-ossifiers have in a strange way colluded to make that term unusable in this context.


magistra 06.29.09 at 8:55 am

On the ‘broken science’ idea, a bit of elaboration. I take science in this sense to mean the discovery of consistent regularities: e.g. if you burn sulphur, you will get sulphur dioxide, with extra conditions put in as you need them: if you burn sulphur in the presence of, etc.

Theories such as Freudian psychoanalysis, Marxism, some forms of ‘social science’ etc have attempted to discover such consistent regularities in human behaviour (either individually or en masse). And they have been rejected as science because the regularities/patterns they have seen are not consistent. Not all boys want to kill their fathers and marry their mothers. Not all of the bourgeoise act in their class interest. It’s also unlikely, because of the complexities of systems that involve human minds, that we will in the near future get scientific theories that can reliably explain how people behave in complex real world situations (rather than simplified experimental worlds). It’s not even clear what variables we should look at to decide why somebody acts in a particular way and someone else does not.

In such cases, if you’re studying humans and their culture, then heuristics (X often goes with Y) are useful, not so much to solve the problem, as to help you think about it in a different way. For example, on my blog I’ve just posted about how an anthropological theory of Mary Douglas has got me thinking about how gay marriage is currently conceptualised. Douglas doesn’t try to prove her theory is universally true, and it probably isn’t, but it has given me an insight I wouldn’t otherwise have had and helped me make what I hope is a plausible argument. Similarly, if a Marxist take on history provides insights into seventeenth century England that you wouldn’t otherwise get, it doesn’t really matter if the same theory doesn’t work for twentieth century Russia, or why it doesn’t work universally. If a Freudian reading of a particular story makes a literary critic see something they hadn’t seen before in the text, then why worry if it isn’t helpful for another book? There isn’t one correct answer that a historian or a literary critic can get anyhow, so it doesn’t in that sense matter if the theoretical tool you use is scientifically flawed.

So in those kind of ways, yes, insights from EP could in theory be useful. The idea that humans are particularly concerned to detect cheats/freeloaders is an interesting way of thinking about human societies even if it proves not to be psychologically accurate. But that doesn’t get away from the sexism problem.

The sexism of (capital letter) EP isn’t about widely accepted differences between the sexes, such as physical strength. It’s specifically about mental differences, where the evidence for consistent differences (not removable by training etc ) is weak on almost all attributes. There is room for rigorous studies of differences and similarities, but EP is unlikely to provide that. Instead, it tends to assume mental differences in the sexes (because so much of its model is based on sexual selection) and then try to demonstrate them, very often by dubious means. See for some particularly awful examples Bad Science and this article, which bases a theory of differences between sexes in foraging on behaviour in a US farmers’ market (!)


JoB 06.29.09 at 9:37 am

magistra, you’re right of course; mental productions of humans are very weakly linked to this, that or the other genetic predisposition and very strongly linked to education. The problems in EP are that it overstretches its claims. Some habits can be shown to be inherited differentially – but these habits are not at the level of mental productions which clearly did not evolve at same pace as the evolutions of the hereditary factors. But that still leaves open the question whether there are habits that have evolved differentially which you would count as ‘sexist’ if pointed to. So it comes down to defining ‘mental’ and going through tedious examples, some of which might engender emotions. But what seems abundantly obvious is that there is no reason why in art, in society, in politics, in science there should be any difference between the sexes (at least not if it can be conceeded that it is good for both sexes to eliminate typical male-habit enforcing rituals from those scenes). Not that there is anything wrong with male habits, they should just not be a sole determinant of who gets a shot at calling the shots.


Tim Wilkinson 06.29.09 at 12:48 pm

Magistra yes, don’t disagree except I suppose I’d just be more categorical in insisting that use of theoretical structures whose scientific application is discredited must be eliminable – and that failing to eliminate it brings with it the risk of unexamined use of vacuous or misleading jargon. Relying on them in ‘abductive’ reasoning (hypothesis-selection) in a wider context of diverse perspectives being brought to bear is I suppose OK.
And I’d want to be quite clear that by There isn’t one correct answer that a historian…can get anyhow we don’t imply that nothing counts as an incorrect answer.

it tends to assume mental differences in the sexes (because so much of its model is based on sexual selection) and then try to demonstrate them, very often by dubious means I’m not trying to defend the just-so-story, Venus/Mars type guff, but given the foregoing, I’d say one has (tediously enuogh) to concentrate on the dubious means bit; there is no shortcut to wholesale rejection. It’s a bit like my pet subject, so-called conspiracy theories, in that regard.

More generally, I’m interested in the fairly obvious, but AFAIK underexamined, implications of EP for the existence of innate ideas, or instinctual cognitive content. How, to choose an example without endorsing its content, would ‘providing for the family’ be encoded in genes (presumably along with a network of other concepts) in such a way as to ‘trigger’ behaviours such as getting a job in an office. And what JoB said.


JoB 06.29.09 at 2:37 pm

Tim, I can’t give you examples but I don’t think it’s underexamined (most of it unsurprisingly is similar in humans to what can be perceived in other animal species). I do think rather that it has been slightly overexamined. There clearly is no support for anything language-based coded in a genetic way for the simple reason that language appeared far too recently. Language is more the context in which we live than it is an encoded predisposition (although, conceivably, if would be sexist enough for a million years or so, we would wind up living in pre-encoded macho-universe that would perfectly suit the sexists of today – not to worry, I guess, because the whole race is to self-destruct before the first hundred of that million years are over).


Tim Wilkinson 06.29.09 at 3:11 pm

I fear lack of examples/refs is a bit of a problem.

Do you suppose that cognitive content is necessarily dependent on (public!) language, rather than say on some sort of pre-linguistc archetypes? And did language definitely appear too recently in any case? It might be, again FAIK, that primitive spoken language was around for a long time before any kind of writing, and possibly before some kind of rapid explosion in vocab and grammar, which latter is a plausible scenario I should have thought. Certainly it must be the case that the physiology required for speech has been around since before the retrospective evolutionary short run, if that makes sense.

Not that I have any commitment to EP – especially not the crap that abounds uder that banner. And if it’s worthwhile to say so, I’m not in favour of sexism! (Not consciously anyway.)

Come to think of it, there has been some stuff in philosophy of language about recognitional capacities for natural kinds – and some – I think related – stuff I can’t remember about ‘proper function’ – (Ruth Barcan?) Millikan was a proponent IIRC, but that was probably at least about 10 years ago and I would have to look it up.

Waiter, this conversation isn’t very good. Bring me a big bowl of specifics.


Bill Benzon 06.29.09 at 4:03 pm

Language as we know it is usually dated to roughly 40K to 100K years ago, though we don’t really know. There was probably some kind of proto-language before that, though just what, we don’t. There’s an extensive literature on this; look for Derek Bickerton, Philip Lieberman, Terrence Deacon, also Merlin Donald (Origins of the Modern Mind), and Stephen Mithen (The Singing Neanderthals (my review here)). Writing is 5K to 8K years old.


JoB 06.29.09 at 4:12 pm


AFAIK (& I don’t have time to be waiting tables, LOL), innate natural kind recognition, and the reflex-part of language is well supported by evidence (across animal species by the way).

But developed language (anything at or above the level of 1st order predicate logic) is, although heavily defended by Chomsky and Fodor and some EP, quite not-innate. The original argument from (I forget Chomsky’s label) lack of public input for children to construct the mechanism is, afaik, abandoned by most. So yes, cognitive content (anything requiring at least 1st order logic) is dependent on public language and public language of that complexity has emerged relatively recently, too recently for genetic adaptation to it.

For sure, I’ve not tracked recent publications and would also welcome some hencetoforth silent waiters to check on our table ;-)


Tim Wilkinson 06.29.09 at 4:33 pm

BB: thanks (and for specific refs). Unnecessarily tentative, I suppose, in assuming that speech would precede writing by a long time. 100K would certainly put it on an evolutionary timescale, wouldn’t it? Not that I’m ready to concede that innate concepts or archetypes need depend on spoken language anyway.

Singing Neanderthals is a rather haunting image, perhaps because of subsequent extinction/their non-human (thus animal?) status…

BTW I conflated Ruth Garrett Millikan (‘Biosemantics’) with Ruth Barcan Marcus – (only somewhat related: ‘Direct Reference’)


Tim Wilkinson 06.29.09 at 4:50 pm

JoB – not sure [edit – actually, I have no idea] what reflex part of language is, but in any case, still don’t see why language is deemed necessary for something general concepts usable in practical reasoning – which I suppose might together be described as instantiating/depending on predicate logic, but surely needn’t, and in any case wouldn’t be an example of language development.


Bill Benzon 06.29.09 at 5:27 pm

Don’t know quite what you mean by practical reasoning, Tim W, but chimpanzees can create and use (simple) tools and, so apparently, can crows. Primate social relations are quite sophisticated and complex.


JoB 06.29.09 at 5:59 pm

Oh! (as to signal surprise)

Auw! (in response to pain)

Anyway, not quite willing to dig me a hole on this one.

Practical reasoning: not quite as Bill takes it. Actual explixit reasoning. Recognizing a spider as dangerous isn’t what I intended. Realizing that not all spiders are dangerous, whatever our instintive reaction, is, I’d say, something kinda hard without language in a developed sense

But as Bill said it: animals are quite continuous with humans except for that one fact of language.

100K years, nope, don’t think it would suffice but would be excited if evidence quoted otherwise. Also, the language of 8K years ago did develop afterwards.

Bill, are you – by any chance, a biologist?


Bill Benzon 06.29.09 at 6:25 pm

Nope, I filed a dissertation on “Literary Theory and Cognitive Science” in 1978. I was in the Department of English at SUNY Buffalo at the time, but the much of the dissertation was a technical exercise in knowledge representation. The only one who really understood my dissertation was the outside reader, David Hays in the Linguistics Department. He was a computational linguist. As for just what I am, it’s hard to tell.

Not sure why you’re concerned about 100Kya and 8Kya. Of course language has continued to develop since writing first merged.

FWIW, the literature on folk taxonomy reports that languages of pre-literate cultures seem to lack words corresponding to “animal” and “plant,” though their syntactical systems have markers for those categories. But without such words it would be pretty difficult to utter propositions involving universal quantification over those categories. AR Luria has a little book that bears on similar matters, Cognitive Development (Harvard UP). It reports results of a study Luria conducted back in the 1930s on cognitive differences between pre-literate and literate peasants in Uzbeckistan.


Tim Wilkinson 06.29.09 at 7:14 pm

#35 should read ‘something like general concepts’, of course…

Actual explicit reasoning.
Requiring it to be explicit seems to be stipulating your way out of the question – or if not, the ‘explicit’ needs some explanation – maybe something like ‘not inchoate’, but that doesn’t seem to clarify anything much. There are undoubtedly epistemic issues in ascribing particular thoughts/sense/meanings/intentions in the absence of language – but IIRC, if the likes of Quine, Davidson are to be believed it’s plausible that language doesn’t really get rid of those anyway.(?)

I’d say that a chimp getting hold of a stick to knock down a mango (A mango? In Africa?) might well count as manifesting practical reason, possibly involving concepts of ‘high’ etc*:

Get food
Food up high

Get up-high-thing

Get up-high-thing
Get up-high-thing -> hold stick

Hold stick Q.E.D.

kinda hard without language in a developed sense – kinda hard is OK. No rush. The epistemic issues haven’t really been resolved by say Wittgenstein’s conceptual turn and the convenient behaviourism-but-not-behaviourism of beetle boxes etc. And certainly not with any ‘private language argument’ I have ever encountered. (-But zere is nussink to resolve! Vhy can’t you see zat?!! -Yeah, yeah, calm down, Ludwig – put it back in the grate.)

Maybe something importantly relevant has passed me by, I dunno. But if any kind of ‘innate ideas’ hypothesis were permitted into the argument, I’d say it would pretty well undermine the merits of arguments from indeterminacy that purport to establish the dependence (or what you will) of higher cognition (or what you will) on language. And it is surely possible (and must have been done) to test for the possession of (ability to behave as though possessing) a concept. Didn’t they do something with a sticker on an elephants head and a big mirror that was suposed to prove something-or-other?

100K years, nope, don’t think it would sufficeThere’s obviously no absolute timescale for evolution beyond one generation (arguably I suppose not even that), but obviously a lot of factors would be relevant – how quickly a mutation could spread throughout the population (or rather a population-successor arise that includes only descendents of the mutant – no doubt still an oversimplfication). I’m afraid my pre-history is so shockingly bad, I don’t even know what the current guess is re the latest time at which there was a reproductively contiguous human population. Someone must know. Maybe I’ll Wiki it. Seems a safe enough topic.

[* Examples are merely illustrative and no warranty, express or implied, is made. Actual chimps’ cognitive processes may differ from those shown.]


JoB 06.29.09 at 7:15 pm

Bill, interesting – I once filed a dissertation on “Do humans think?” – answering in the negative as far as our brains having any capacity for direct deductive reasoning – but it failed to generate the upheaval I imagined it would have – it did earn me almost top marks but maybe that was mainly because the psychologist jurors were impressed by references to Kyburg, Eddington and much more obscure logicians’ stuff ;-)

Why I’m concerned on 100 Kyears & stuff? Because I thought the debate was – inter alia – on whether such length of time would be enough for innate developed language to develop. Not so much interested really, I don’t think the idea of an innate language module is even consistentlu conceivable.

I don’t know where you want to go with your ‘FWIW’ but it is quite possible that there are genetic predispositions for concrete animals without concepts like “animal” whilst language can develop to such syntactical markers.


Tim Wilkinson 06.29.09 at 7:22 pm

Oh yeah, excuse my ignorance (read: this is a genuine question, not criticism), but how does they seem to lack words corresponding to “animal” and “plant,” though their syntactical systems have markers for those categories work? Most intriguing – what are those markers?


Tim Wilkinson 06.29.09 at 7:33 pm

I don’t think the idea of an innate language module is even consistently conceivable.

Waiter! This is quite indigestible.


Bill Benzon 06.29.09 at 8:12 pm

I don’t know where you want to go with your ‘FWIW’ but it is quite possible that there are genetic predispositions for concrete animals without concepts like “animal” whilst language can develop to such syntactical markers.

“Concrete animals” is tricky. Strictly speaking, all you ever see are individuals. Presumbably you’re interested in the most primitive grouping of individuals into some “natural” kind.

There’s been quite a bit of work on that – it’s called ethnobiology – over the last 40 years or so. It seems that languages “start” with what is now called the “basic” level and then move up and down. So, horse, dog, and pig would be basic level among concepts. Go down a level and you get, e.g. collie, beagle, etc. Go up a level and you get beast (aka “any sized four-legged furry creature with a tail”). You don’t always find a “beast” term and, as I’ve said already, you never find an “animal” term.

Tim, as for an example, since I can’t speak any of those languages nor do I study them in the way a linguist would, I don’t keep such things in mind. Check out Brent Berlin, Ethnogiological Classification (Princeton), which also says lots about basic level terms, etc.

As for an “innate” language module, back in 2003 Science published an article by Fitch, Tecumseh, and Chomsky (or maybe it was TF & C) which addressed the biological basis of language. I’m afraid that all that was left of innateness was recursion; everything else was allowed to be derived from other stuff. This kicked off a disput in which Steven Pinker and Ray Jackendoff took a different position (I forget just what at the moment, probably more innate stuff). I think this dispute took place in the pages of Cognition or Cognitive Science; it went back and forth several times. You should be able to find articles at Pinker’s web site, which should turn up upon googling.


Tim Wilkinson 06.29.09 at 8:37 pm

I meant – what kind of things would those markers be if not words?


Bill Benzon 06.29.09 at 9:07 pm

Well, in English “ing” is a marker that gets attached to verbs to indicate the progressive tense while “ed” is a marker that indicates past tense, thus “jumping” and “jumped.” Other languages attach such markers to nouns to indicate the relationship between the noun and the verb: Is the noun the subject of the verb or the object, and if the object, is it direct or indirect? Plants cannot coherently be subject of a number of verbs that accept animal subjects just fine, e.g. run, eat, see, etc. Plants can be subject of verbs that could not accept mere inanimate objects (stone, cloud), e.g. grow, live, die. This sort of thing gets philosophical treatment under the rubric of category mistakes. Well, those markers are there to indicate categorical suitability and role in syntactic constructions.

I’m told there are primitive languages that make Latin seem grossly deficient in the “marker” department. English, by contrast, has rather fewer such markers than Latin, but gets the job done by independent words and by syntactic order.


Tim Wilkinson 06.29.09 at 10:55 pm

So markers in general would be parts of words (or gaps where such might have been, + complications for irregular verbs etc…) that reflect/dictate (some restriction on) the range of possible syntactic roles. Although the examples of tense, for one, seem to involve semantic properties, possibly without any syntactic implications.

You could, presumably, correctly ascribe a category-concept in the absence of such markers, based directly on syntactic role (and the accompanying semantics). I suppose having strict categorial markers would tend to inhibit use of metaphor – and thus of new usages that arise from widely accepted metaphor…but that’s (even further) off-topic.

More on-topic, it strikes me that explicity framing syntactic properties as restrictions on possible syntactic roles suggests at least in principle a way of distinguishing innate from environmental aspects of language – an innate restriction means that however the environment is/were altered, the restricted structure will/would not arise.

Anyway, still no-one has ever shown me to my satisfaction that concepts must be linguistically-based, nor that they can’t be innate.


john c. halasz 06.29.09 at 11:05 pm

In order for something to be perceived, it must be categorized. (One doesn’t perceive a loud metallic blur rushing by; one perceives a fire engine racing toward a presumed fire). In order for there to be categorization, there must be a system of categories, (i.e. more than one category), which itself implies a system of differences, which make a difference, (presumably of behavioral import). Primary-perceptual consciousness as a neurally embedded and generated system in a simultaneous integration of the senses into a presentation of the environment connected up through on-going re-entrant loops with a value-weighted categorical memory: it’s a “remembered present”, following Bergson and Whitehead. A “concept” would be any relation between percepts that is not itself a percept. So defined, concepts are already involved in the structuring of any rudimentary primary consciousness system. They needn’t be “rational”, but they would tend to be selected insofar as they confer some sort of generalized control over environmental events or “objects” in the course of behavioral interaction with the environment. But categories and concepts needn’t be innate or fixed genetically by the physiological structure of the brain. They can evolve, change, and even be acquired ab ovo through neural experience, depending on the plasticity and processing capacity constraints of any given neuro-physiological system. (Central nervous systems emerged and evolved initially for purely physical and physiological reasons or selection pressures, to integrate the metabolic functions of the organism and mediate them with behavioral interaction with environmental events. There is no clear line between the physiological, the behavioral and the mental here, and what exactly is identifiable as “mental” minimally, as an emergent property or capacity of the system-processing as a whole, is a ticklish question. But rudimentary consciousness would presumably initially have emerged as conferring a large selection advantage in terms of the integration of perception and motility for organism with such, say, a mouse, versus organisms without it, say, a lobster, which would have senses and value-category weighting, but no simultaneous integration of them. But consciousness itself, which can’t, partly for the reason of unclarity above, be identified with mentality per se, is itself a selective system, selecting for salience amongst the excess of information its total awareness generates).

Now when a given neuro-physiological system develops the emergent capacity to map its own processes and through re-entrant loops to re-categorize its own categorizations, then it begins to take on semantic capacities. (There is some considerable evidence of such attributable to chimps). An organisms so endowed would begin to be able to treat percepts as signs, (“where there’s smoke, there’s fire”), and draw behavioral inferences from them. Henceforth increasingly directed learning process involving concept acquisition can come to be inter-nested with the development of neural experience, relatively unchained from fixed instinctual responses. When such percepts-as-signs can be recognized as communicational signs in behavioral interaction with con-specific organisms and come to acquire a good deal of flexibility in more-or-less complicated behavioral interactions, then there is the beginnings of animal sociality, as opposed to instinctually fixed gregarious behaviors, and, arguably, the beginnings of group selection, contrary to the traditional taboo on such among biologists, as violating the rule that selection operates solely upon the fitness of individual organisms. Learning and communication and the increasingly flexible behavior they underwrite co-evolve with innate endowments and instinctually fixed or chained behaviors, so as to become inter-nested with them, such that the evolution of the latter come to “suppose” the former.

Now human beings didn’t suddenly develop language and thereby become sociable. Rather language could only have possibly emerged on the basis of a long evolution of animal sociability, going back to the earliest primates at least, and involving increasingly complex neural/mental capacities, which, I’ve no doubt, sociality itself increasingly drove the evolution of. Now not only is language an immensely complicated and multifaceted set of issues, but the relation between language and symbolic thinking is one of those impossible chicken-and-egg questions. But I think it’s a reasonable surmise that language emerged on the basis of a prior analog system of animal communication. In such systems, there is no syntactic “aboutness”, but communication occurs entirely as and through a specification of a relationship in a behavioral interaction between organisms, such that the relational hierarchy of the group as much serves the needs of communication as communication serves the maintenance of the hierarchy and cohesion of the group. (It’s an interesting speculative question nonetheless how much learned informational “aboutness” could be transmitted through such relational communication in conjunction with mimetic behavior and learning). As an increasingly complex and differentiated set of activities and relationships developed to be mediated by the relational hierarchy/analog communication system of the animal group, such an analog communication system would tend to undergo a crisis of over-complexity and tend to break-down. The “solution” would be a re-differentiation of the system though the digitalization and recombination of initially analog vocalization signals, through which, in a perhaps gradual but self-accelerating way, phonemically coded, syntactically structured and semantically self-stabilized human natural language systems would come to be organized. But such language is both a digital and an analog system, (which the biases of linguists tend to overlook), and that analog-relational dimension is still very much a part of our language and how it “means”. The upshot here is that while, of course, human organisms must have the sorts of brains (and vocal tracts and ears) that can handle the processing of language, language is not an entirely intra-cranial matter, nor the projection and extension of pre-existent “mind”. The possibilities for re-combinant symbolism in behavioral interaction and communication across the world massively amplify the scope and range of the semantic articulations and discriminations that individuals can make, and much of what we would be wont to attribute as human “mind” is an internalization of such communicative processes across the world. Which is also to say that “human nature” has become in large measure cultural, since the complexity and plasticity of human social interactions have largely evolved to break down fixed instinctual endowments, leaving us organically under-determined and insufficient to adapt and survive on the basis of a pre-given natural endowment. Human beings have evolved to require cultural structuration to channel and evoke their biological needs and endowments. The old opposition between nature and culture is confused, precisely because it misses that excluded third, wherein nature and culture inter-penetrate: culture is part of human biology.


Bill Benzon 06.29.09 at 11:59 pm

You could, presumably, correctly ascribe a category-concept in the absence of such markers, based directly on syntactic role (and the accompanying semantics).

The issue is how you recognize syntactic role. That’s what conventions of word order the markers are for. Sentences are not simply bags of words. The involve relations between those words. Consider this linearized bag of words: Bat Fred ball. What is being asserted? Consider these three possibilities (among others):

(1) Fred hit the ball with the bat.

(2) Fred hit the bat with the ball.

(3) The bat hit Fred with the ball.

We can rule (3) out on semantic grounds, but if this word bag occurred in a SF or fantasy story involving giant super-intelligent bats from another dimension, then who knows? Both (1) and (2) are semantically possible, though (1) seems rather more likely. But if Fred were a pitcher, then (2) would be possible, though as bit odd.


Bill Benzon 06.30.09 at 12:19 am

Whoops! Missed something. The linearized bag of words should include “hit” somewhere, thus: Bat Fred ball hit, or Hit bat Fred ball, or Fred ball hit bat, etc.


Tim Wilkinson 06.30.09 at 12:36 am

john c. halasz @47
Bit too much and a bit too dense to wade through in detail at this time of night. To give a few (negative – so much easier) responses:

Not very keen on your digital/analogue metaphor if you’ll excuse my saying so.

Nor A “concept” would be any relation between percepts that is not itself a percept. 1: says who? 2: the relative clause appears redundant.

the relation between language and symbolic thinking is one of those impossible chicken-and-egg questions
But Darwin solved that question. Has no-one else noticed? It’s almost literally definitive of his great achievement. In any case, I suspect the appearance of interdependence rests on a linguistical bias in the specification (or the chosen interpretation) of ‘symbolic’ thinking.


Tim Wilkinson 06.30.09 at 1:19 am

I should probably @48 have said ‘structure’ rather than ‘role’. I appear to have inadvertently given the impression of being a trifle hard of understanding.

What I meant was that in languages without category markers, you could infer from the fact that certain combinations were or weren’t permitted by convention that there was some such category – in the same way as you would initially infer the meaning of a category marker.

On reflection I suppose category markers would only limit the use of metaphor, flights of fancy and (as you suggest) Sci-Fi if they were undetachable – which I unjustifiedly assumed they would be.

Given the addition of ‘the’,’the’ and ‘with’, Newton (or maybe Einstein) might be happy enough with ‘The ball hit Fred with the bat’.

That’s what conventions of word order the markers are for. I genuinely can’t work out what this should say. I’m guessing ‘the’ should be ‘and’.

Anyway that’s enough hijacking for one night.


Bill Benzon 06.30.09 at 2:09 am

Yes, “and.”

‘The ball hit Fred with the bat’.

Did Newton or Einstein assign agency to inanimate objects? The physics wouldn’t require it I wouldn’t think.


john c. halasz 06.30.09 at 3:28 am

Tim Wilkerson@ 50:

Different time zones, no doubt, but…

No, there’s no redundancy in the relative clause; it was plain English, and an organism can respond sequentially to different percepts, -(too hot, too cold),- without the imputation of any relation between percepts, whether innate, acquired, or directively learned.

As to “who says”, I’m following Gerald Edelman’s neural theory, with my own convolutions or elaborations, in the first part, Arnold Gehlen’s ,- yes, another ex-Nazi,- in the concluding coda about the organic insufficiency/plasticity of “human nature” requiring cultural structuration.

As to the analytic/digital distinction, it’s no “metaphor” and I’m sorry if it doesn’t accord with your mere taste. But, it’s basic, among other matters, to general systems theory, which most especially was developed to deal with the modeling of biological organisms, though it’s, by no means, a simple distinction, as the establishment of a boundary, which must occur for a system/environment relation to take hold, let alone any internal differentiation within a system, always involves a digitalization of that boundary. But I’ll state, quite dogmatically, that brains/neuro-physiological systems must be primarily analog pattern-matching systems rather than digital computational devices for good biological/evolutionary reasons.

And I have no idea what you might mean by Darwin having solved these issues once and for all, as if he had conceived of them within his particular quest/problematic entirely. At any rate, the main burden of my comment/post is that there is a large middle ground between biologically innate/genetically pre-determined and entirely open-ended/environmentally “determined”, and most of the realistic answers would likely be found, though variously distributed, within that “excluded middle”, which mistaken framings of the issues construct. I might as well say that Whitehead’s “reform” of focusing on processes rather than substances resolved the mind/body problem once and for all, though I think it was a constructive suggestion. At any rate, even within a post-Darwinian, evolutionary perspective, the appeal to 17th century rationalist “innate” ideas, as per Descartes, Leibniz, and those Cambridge Platonists that Chomsky was so enamoured of, as equated with biologically innate causal processes, would hardly suffice, as the causal explanation of a capacity doesn’t forthwith translate into a causal determination of that same capacity, by plain logic, though it might entail structural constraints or limits on that capacity. You’d think that a step-wise approach to causal explanations, as implied by a “Darwinian” evolutionary approach, might at least hint at why more generalized, less narrowly determined capacities emerge. And at why natural language might imply a semantic “explosion”, as compared with a strictly biologically self-enclosed semantic capacity.


Tim Wilkinson 06.30.09 at 7:43 am

Mornin all. Sorry about flippancy chaps, got a bit thread-death-happy.

BB re: ‘The ball hit Fred with the bat’ – Did Newton or Einstein assign agency to inanimate objects?
Not AFAIK unless it featured in Newton’s mystical alchemy. Perhaps stretching proper usage a bit, I was allowing ‘hit with’ as an agency-neutral ‘impacted via’. Thinking equal-and-opposite reactions/Relativity.

JH –
To revisit #47: A “concept” would be any relation between percepts that is not itself a percept:
But this 1. is too idiosyncratic to be recognisable as a conception of the concept |concept| – hence stipulation-quotes I suppose; 2. Would entail that, say, percept a’s occurring earlier than percept b counts as a concept; 3. Seems to refer to percept tokens rather than types, so leaves concepts (or “concepts”) unrepeatable. And re: there’s no redundancy in the relative clause – so a relation between percepts might itself be a percept (rather than at best an object of perception)? If you would say I’m reading too literally, I’d say you’re not writing literally enough.

On which subject, digitalization and recombination of initially analog vocalization signals sounds pretty metaphorical to me. What digits? A no-less-apt metaphor might be the move from base 1 (like Roman numerals 1-3) to some higher base.

Darwin having solved these issues once and for all
Not those issues – just the one about all chickens coming from eggs but not vice versa. (Or if you stipulate ‘egg’= ‘chicken’s egg’, only vice versa.) It does seem to me odd that no-one seems to have pointed this out as a genuine achievement of Darwin’s theory – nor indeed as a mere curiosity or anything else.

A navel full of coarsely chopped logic really sets you up for the day, I find. Or, the fourth ‘S’ of the morning routine – semantics.


Bill Benzon 06.30.09 at 10:01 am

…analog pattern-matching systems rather than digital computational devices …

Analog computation is a perfectly coherent notion, as is digitical pattern matching.


JoB 06.30.09 at 11:35 am

Ouf! I thought I’d had to respond to Tim’s waiting analogy by saying that somebody calling for a waiter once too often might have crossed that thin line where he’s utterly unaware that he starts making a nuisance of himself. But John c h rescued me of that. Awesome post, John! Thanks for that. I hope I can come back to you during daylight in Europe.


JoB 06.30.09 at 12:58 pm

John, Tim, Bill,

The idea of the modular mind with innate modules adapted to specifically human skills such as language, deductive reasoning and morality has dominated cognitive science for some time – an excellent book was written by Jerry Fodor (where he explicitly makes the link with phrenology by the way whence one can link back to the original topic of the thread). Chomsky, Pinker, and many others here referenced joined in that discussion. Modularity and computationalism – not the analog variant if you insist to use the word computation for this – are intimately connected, as is the notion of the brain as an information-processing device, a computer. The link with EP is also intricate because, wrong as I think they are, they were honest people and needed to have a plausible account of how these modules could have been perfected for such specific needs.

This consensus has been attacked from two sides.

It has been attacked by denying the central tenet of modularism and computationalism: ‘there is lack of information in our childhood phase to account for the exceptional human competences of children & adults; hence there has to be a specific genetic predisposition’. The attack centers on (see John above) the ability to account for the emergence of these competences by a cultural (and basically learned and therefore non-discriminative) progression making use of only a basic number of functions (concepts, categories, association, probabilistic trends and vocalization of words) that are (except the latter) not unique to the human race. Advances in neural networks, or analog computers if you insist, have made this claim very plausible.

The other attack is on the consistency of the idea itself. It basically is an argument against some kind of dualism which is unavoidable in the computational view. In essence it isn’t very far from what somebody like Dennett says on the matter (and where he is endebted to a scientific outcast by the name of Julian Jaynes). But the real work has been done by Quine & Davidson on the real irreducibility of language to a single individual or a single brain; on the inconsistency of what we understand to be expressible in a definitive way without interaction. By the way, progress of AI is mostly in the direction of creating modules that ‘learn’ from each other; not in programming, not even with large assistant neural networks, specific functions emulating the external powers of reasoning of human beings.

(I don’t imagine for a moment that I was a good waiter here. I’m not in the knock-down business)


Tim Wilkinson 06.30.09 at 1:57 pm

JoB Don’t blame me for dragging out the waiter reference (not an analogy). I used it once to note that both of us were failing to provide specifics. It became a running gag (or sore) once it was used twice, so I rather unfunnily used the motif again in preference to being a bit more explicit about the shortcomings of a sentence which seemingly claimed to be incomprehensible.

Anyway I now have absolutely no idea what this waiter is supposed to symbolise, in relation to this ‘knock-down’ business or otherwise. A dead parrot reference would be more appropriate here I think.


JoB 06.30.09 at 3:50 pm

Tim – sorry, I overstated, I thought you expected knock-down arguments or final proof calling for your waiter. I meant to convey my belief the fun was out of it by trying and failing at an own attempt at fun. Peace.


john c. halasz 06.30.09 at 4:06 pm

Tim Wilkinson @54:

Sorry for misspelling your name before. (But then, ya know, all lymies look alike to me). But I fail to grasp what intelligible points you making in the nits you pick. 1. The definition of “concept” is deliberately minimal, even privative, aimed at how concepts might be physically instantiated, therefor possibly innate, but also acquired through behavioral learning. 2. just reminds me of Strawson famously declaring Kant guilty of a “non sequitur of numbing grossness”, when Strawson plainly mis-read Kant. Temporal succession, at any rate, would not plainly count as a sense percept, would it. 3. Well, percepts are events/tokens, except that they require categorization/memory to occur, and hence must have already occurred. I didn’t invent that paradox; it’s just there. Concepts are clearly tokens, hence repeatable. Any structuring of percepts, innately or behaviorally, tends to yield “concepts”. But then it’s a standard saw that any system needs to “de-paradoxicalize” itself, if it is to at all get going. And I have know idea how a relation could be an object of perception. That sounds like the sort of “philosophy of mind” that old Ludwig was attempting to undo. One doesn’t, e.g., perceive one’s self or one’s own thoughts: that’s just a mis-use of the word/concept “perceive”.

Digital means operating and processing “information” through discrete units, such as phoneme-and-morphemes, or numbers. Analog means operating and processing “information” without units and yes/no, either/or decisions, just a continuum of more or less, similar or different, like a wave function. (And it’s very tempting to talk of neural processing by means of a wave metaphor or analogy).

But then I have no idea what, if any, position your arguing for. If we’re arguing over a dead parrot, then who bought it?

For the rest JoB has straighted out any dispute about “computation”.


JoB 06.30.09 at 7:03 pm


I’m hard-core: I don’t even buy analog ‘processing’ as it still would entail something in the head that’s somehow directly representing something outside of it. It’s tempting, & I’m sure to cover ground we need to revert to that way of speaking, but it’s wrong (and not innocently wrong in an academic way but really damagingly wrong day-to-day – an example of which is the EP-stuff and sexism with which this started).

My inspiration was with Gibson – for anything sophisticated one just cannot isolate the process of perceiving from the activity of perceiving (sorry, too hermetic).

The interesting bit really is to discover how with such limited means ( i.e. association, correlation and vocalization) we have gotten to the incredibly expressive languages & how there is no limit to the creativity that can be expressed in it – not even the limit of recursion. It should not remain a miracle or just a Bergsonion drive. That’s where I’ve tried to study, before the reality of keeping my current standard of wealth for the kids kicked in and I had to start frequenting blogs ;-)

I’m sure the next Darwin will be the one cracking the cognitive science mess.


lemuel pitkin 06.30.09 at 7:13 pm

JoB and Bill Benzon,

Both of you are making exceedingly interesting arguments. Any recommendations for non-blog reading on these questions?


JoB 06.30.09 at 8:23 pm

lp, thanks (I’m all blushing now, & for an hour or so I could only type ‘Duh, Duh, Duh’ feeling all gooey in the stomach) and at the risk of being too tacky (or overly ironic): I enjoy your posts as well so, thanks.

I fear I can’t give you direct reading material (non-blog at least, since I do have a blog) as I assembled this over the years. Except for the usual – and already named – suspects there’s Kyburg, Eddington and Grice (for technical stuff), Quine & Davidson (for all of the theory), Habermas (links to morality), Gibson & Bergson as inspirations, works on bounded (or what’s it called,) rationality I guess and much more. The peak of modular computational model is fairly recent and although being abandoned I don’t know of an individual that has undone it and put something alternative in place.

Anyway, hope you weren’t ironic, if you are really interested, I can send you my stuff.


Bill Benzon 06.30.09 at 8:58 pm

lp: a bunch of my papers are online at the URL beneath my name. This paper on natural intelligence and this one on cognition are basic. You might also want to look at this one on metaphor as a neural process, which is not so hairy as the other two. Those papers have lots of stuff for you to cite.

You might also want to look at John von Neuman’s little 1958 book on The Computer and the Brain. Don’t let the ancient publication date fool you. It’s an insightful discussion of what’s required to get physical matter to perform computations.

And then there’s my book on music, Beethoven’s Anvil: Music in Mind and Culture. I’d say the most important thing in that book is the argument I make in chapters 2, 3, and 4, about how music is fundamentally a collective process among different people. I make that argument at the neural level, using the neuroscience of Walter Freeman and some stuff on coupled oscillation (fireflies and handclapping). There is a chapter on the evolution of music, though it’s not on the EP line.


JoB 07.01.09 at 8:40 am

Bill, “Mind-Culture Co-Evolution” … great phrase, I put the link in my memory for when I have time to re-enter this line of study; hope I get there, hope you are around. I’d wish I had my stuff on-line, will have to ask my computer friend to host it somewhere; mine is much more technical (from a logical point of view) and at the same time more all over the place. As am I ;-(

Comments on this entry are closed.