Explanations in the Social Sciences

By Daniel A. Kaufman

I want to sketch some ideas on explanations in the social sciences and how they differ from their counterparts in the physical sciences.  The point is to start a discussion, which I hope will further sharpen my own views on the subject and pave the way for a subsequent, academic paper, as well as a dialogue, with my good friend, Massimo Pigliucci, a philosopher and biologist, at the City University of New York.

Clearly, this is a much-discussed topic, and a number of classic works inform my views here, including Gilbert Ryle’s The Concept of Mind (1949), Peter Winch’s The Idea of a Social Science and its Relation to Philosophy (1958), Jerry Fodor’s “Special Sciences (or the disunity of science as a working hypothesis)”(1974), and of course, Ludwig Wittgenstein’s Philosophical Investigations (1953).  That said, for the sake of brevity – and hopefully, clarity – I will not explicitly cite any of these works, here, though they will loom large, in the background.

____________

1. In the social sciences, we are not interested in explaining/understanding human motor movement, per se, but rather, human action, which consists of motor movements conceived under various intentional descriptions. Thus, in abnormal psychology, we are not interested in, say, the movements of peoples’ eyes and the dilations of their pupils, but in voyeurism.  In social psychology, we are not interested in the tensing and flexing of arms, legs, and hands, but in mob violence.  In economics, we are not interested in the making of ink marks or the pressing of buttons, but in trading practices.  In each case, then, it is the interpretation – and thus, the meaning and significance – of the various, underlying physiological events that provide the impetus for our social scientific investigations, not the physiological events, in themselves.

2. The social sciences seek to identify reasons for the sorts of human actions that I have just described. The question is how we should understand those reasons.  In the physical sciences, the reasons we seek for the behavior of animate and inanimate matter are causes, and beyond the pleasure inherent in discovery, our main reason for wanting to know the causes of things is so that we can predict and thereby, exert some degree of control over our material circumstances.  Of course, ‘cause’ is itself a problematic notion, for which there are competing and thus, controversial accounts (whether probabilistic, counterfactual, or in terms of necessary and sufficient conditions), but I trust that we all have enough of an intuitive sense of what is commonly meant by the term that we will be able to contrast the idea of reasons as causes with the very different conception of reasons that I am going to claim are the bread and butter of social scientific accounts.

How should we understand the sorts of reasons sought in the social sciences?  Imagine that we are told by a clinical psychologist that the reasons for a particular young man’s voyeurism are the feelings of sexual shame and the beliefs concerning his own sexual inadequacy that he acquired, over the course of an abusive childhood; or that a social psychologist says that the reasons for the mob violence we have just observed lie in a list of longstanding, unaddressed grievances on the part of the rioting group, combined with the manipulative rhetoric of their charismatic and demagogic leader; or that an economic analyst tells us that the reason for this particular investment firm’s current trading practices is that its senior partners’ share the belief that we are about to enter a bearish market.  Do these sorts of reasons provide causes, in the sense of ‘cause’ meant in the physical sciences?  I’m not sure and am increasingly inclined to think not.  (More on this a bit later.)

Something that they clearly do is render the behavior in question intelligible in a certain way, by providing us with a series of relevant rationales.  If the identification of lawful cause-effect relationships makes it possible for us to predict future physical events, being given a relevant rationale for something someone has done makes it possible for us, in a sense, to infer it; to see it as rational.  Even when the behavior in question is aberrant and (in another sense of the word) completely irrational – such as that of the voyeur, whose behavior is paraphilic – being provided with reasons of the sort we find in the social sciences enables us to see that the behavior makes a certain kind of sense, within a specifically drawn context.  (It is worth noting, in this vein, that social scientific accounts inevitably require a sensitivity to specific contexts that it is the ambition of genuinely causal accounts to abstract from.)

3. The understanding that results involves the ability to fit the behavior we see in our fellow human beings into a rational story or narrative that is necessary if the human world is to make sense, and it is important to us – very important, I would maintain – that the human world should make sense. Do causes provide us with this sort of understanding?  Does knowledge of causes render the material universe intelligible, in the way that knowledge of rationales renders the human world intelligible?  I don’t think so, but can only give the barest sketch of why, here:

(a) Intelligibility, in the sense of “fitting into a narrative” that I have described, is clearly rationalistic and teleological, and if anything defines the modern revolutions in the natural sciences, it is that they have abandoned a rationalistic and teleological view of nature, in favor of a purely mechanical one.

(b) With every new revolution in the natural sciences, we seem to be moving further and further away from a rationalistic story or narrative about the non-human world, rather than closer to one.  Evolutionary biology tells us that the purpose we perceive in nature is illusory, and quantum mechanics casts doubt on whether classical laws of rationality, like bivalence and the excluded middle, hold.  Thus, the natural world makes less and less sense, rationalistically and teleologically-speaking, as our scientific understanding increases, not more and more.

(c) It is precisely because causal explanations do not render the world intelligible, in this sense, that so many people seek to embed the causal, scientific picture of the world into a larger, supernaturally-infused picture (religion) that will provide that sense of the reasonableness and purposefulness of the non-human world that modern science denies.

4. The theist and the eliminative materialist are thus, mirror images of one another: the theist can’t stand the thought of a material world that cannot be construed as reasonable and purposeful, and the eliminative materialist can’t stand the thought of a human world that can be so constructed.

5. The remaining question is whether the sorts of reasons given in the social sciences – that is, intentional reasons – are also properly deemed causes. Beyond rendering human actions intelligible – beyond helping us to see why peoples’ myriad behaviors make sense – do intentional reasons (what I have been calling “rationales”) also give us the causes of human action, such that we can accurately call the resulting accounts “explanations,” as the term is meant in the natural sciences?

This question is most commonly addressed in the debate over whether Folk Psychology should be thought of as a science and whether folk psychological explanations are properly causal and thus, properly scientific explanations.  (Folk psychological explanations are cast in the intentional idiom.) This is unsurprising, given that psychology is fundamental, relative to the social sciences, much in the way that physics is fundamental to the natural sciences.  Part of the difficulty of construing intentional explanations as causal lies in their irreducibility to physical explanations, though this difficulty is hardly decisive – Fodor, who is responsible for one of the most well-known critiques of inter-theoretic reduction is also one of the strongest advocates for the idea of intentional causation.  The bigger problem, to my mind, is a Wittgensteinian one that concerns the very nature of both intentional states and actions.

The idea of intentional (“reason-action”) causation is supposed to be analogous to that of neurological-motor movement causation.  We understand what it means to say that a particular series of neurological events caused the fingers of my left hand to curl up into a fist.  The neurological causes are clearly identifiable as discrete, localizable events in the brain that make certain other discrete, localizable events in the hand occur.  The problem is that there is no clear, analogous story that we could tell about an intentional reason and an action.  A reason, whether a feeling of sexual inadequacy or the belief that a bear market is coming, is not a discrete, localizable event; unlike neurons firing, one’s feeling of sexual inadequacy or belief that a bear market is coming, are not things inside one’s head, to which we can point.  Intentional states are inherently linguistic, insofar as they involve our representing states of affairs in various ways, and are thus necessarily rule-governed. For reasons that I am not going to state explicitly here, rule-following is an inherently public activity, from which it follows that one cannot follow a rule (and thus, employ language) from “inside” one’s own mind.  To speak, then, of beliefs and desires and feelings and all other manner of intentional states as discrete, localizable things or events “in” the mind or brain can be nothing but a kind of nonsense or at best, a highly misleading, metaphorical way of speaking.

Actions also are not properly identifiable as discrete objects or events, for they too are as much a matter of interpretation – of how things are represented in the public language – as they are a matter of motor movements.  In the Investigations, Wittgenstein famously asked what is “left over,” when I subtract the fact that my arm goes up, from the fact that I raise my arm, and the answer that tempts us is that what is left over is a prior, mental act of intending to raise my arm.  We’ve seen, however, that such an answer must be a mistake – that we cannot construe intentional states like “intending,” “wanting,” “hoping,” and the like as discrete, localizable, internal events – and that there must be something else that makes these motor movements count as raising one’s hand, asking a question, requesting permission to go to the bathroom (the latter two of which, notice, can also be effected by one’s hand raising).  That something else would seem clearly to be various interpretations that are warranted, given relevant features of the context and the linguistic rules and conventions of the speaking community.  But this means that the action of raising your hand is no more a discrete, localizable event in the world, then your desire to ask a question or go to the bathroom is a discrete, localizable event inside your head.

Summarizing this point and wrapping up this “sketch,” for α and β to be related as cause and effect, at a minimum, α and β, must be discrete, localizable, events.  One trouble with the idea of the intentional causation of action, then, is that neither intentional states nor actions can be properly characterized as discrete, localizable events, and to think that they are is to commit what Gilbert Ryle has called a “category error,” one that is easy to make, for the sorts of reasons that Wittgenstein devoted a good portion of his philosophical work to describing.

__________

I look forward to what I hope will be a rigorous (and vigorous) discussion and to further developing my thoughts on this fascinating and important topic.

Categories: Essay

98 Comments »

  1. But all I wanted to know was, when you talk of something being a language, you might mean something that involves no actual words. I think that is important context.

    But, as I said, I don’t think there are any any discrete localisable events in physical sciences any more than there are when talking of mental events. In physical sciences we can use a simplification which acts like a discrete localisable event and I don’t see why the same thing cannot be done with mental events.

    If intentional states are public, I don’t see this as any bar to speaking of them as causes as long as we are able to make meaningful statements about it. It is not strictly true to say “the fire was started by the lightning strike” or “the billiard ball’s movement was caused by the cue ball hitting it” or “the heat at the bottom of the pan causes the bubbles to form”, but they will do for certain purposes. So to say that a certain thing was caused by a person’s intention to do it might not be strictly true either, but will also do for certain purposes.

    Like

  2. Not trying to be obscure here, but debating causality in terms of static framing devices, words, pictures, etc, misses the actual reality.

    Maps are handy, but they really do only exist as references to the territory. If they become entirely divorced from it, they are meaningless.

    Like

  3. Robin:

    I don’t know what to tell you, at this point. I’ve pretty much made the clearest, best case that I can. If you still can’t see it, then I’m afraid I can’t explain it any better than I already have. Where you see no relevant difference, I see an enormous one.

    Like

  4. Daniel,

    I am not aware that I asked you to say any more. You made your case, I disagree. I don’t think there are any such things as discrete localisable events, except in models which are a simplification of reality. I don’t think that anybody can point to anything at all which is a cause or an effect given that definition.

    That is all. We disagree.

    Like

  5. Hi EJ. I did choose specific phobia for a reason: exposure therapies are more successful. Since I mentioned the idea that causation implies manipulation, the fact that therapies based on particular models are successful is taken as evidence that the underlying models are more likely to be correct. I fully agree that social contexts have large causative cum explanatory effects. I am trying to present examples of explanation and causal reasoning from a particular approach to the social sciences, the quantitative. It underlies your comment that “talking therapies and behavioral therapies actually tend to have the same success rate” – that is, there is something measurable across individuals that is relevant, reliable and replicable in pen-and-paper psychometric instruments like the Beck Depression Inventory (created by the CBT crowd, btw), which we can plug into linear equations and perform statistical tests on, just like physicists (and economists ;)).

    Regarding holistic phenomena like personality, psychologists from Spearman onwards have used mathematical methods to reduce these to multiple quantities (factor analysis etc). The point is not a theory of everything, but whether there are bits we can measure and then understand.

    Like

  6. @ Robin Herbert:

    “Since no such past usage of language or of these words is alluded to, then what could the sceptic possibly be referring to by the “past use of language” or past usage of words, such as “count”?”

    It is alluded to, as a reconstruction of what the rule must have been (hence the use of the phrase “I may suppose” by Kripke when referring back to the rule as originally taught). Part of what this involves is the question of justification of the interpretation of rules, which according to Wittgenstein come to an end with public agreements, since there is no deeper level than this.
    ____

    “But, as I said, I don’t think there are any any discrete localisable events in physical sciences any more than there are when talking of mental events. In physical sciences we can use a simplification which acts like a discrete localisable event and I don’t see why the same thing cannot be done with mental events.”

    This conflates two entirely different things. The descriptions in the physical sciences are neutral, value-less, and non-teleological. They are mechanical, so to speak. Descriptions in the social sciences are about human behavior, and thus inevitably involve explanations that are not simply neutral and mechanical, but are shot through with interpretive features keyed to value-judgments about whatever the particular behavior is taken to be under that particular value-inflected explanation. Given that the natural world is described in neutral mechanical terms and descriptions of the human world are not neutral and mechanical but vary according to the public agreements of a particular community in terms of what the behavior *means*, it is perfectly just to ask whether both describe causation in the same sense. The fact that physical explanations are general and not at the atomic level is completely irrelevant. They still describe causes in a neutral mechanical sort of way, regardless of the zoom factor. Descriptions of human behavior are completely different and thus the notion of cause is, too.

    Suppose one tree is destroyed in New Guinea and an identical tree in New York City, for the same basic reason. The explanation in terms of both will be generally the same. Now, suppose one young person engages in a motor movement in New York and another in the depths of the Amazon engages in the exact same motor movement. In terms of human physiognomy, both will be described as the same. But we still want to know what both mean (in terms of, e.g. anthropology), since human behavior involves networks of meanings, not just mechanical movements. If the first person’s movement is dancing at a wedding and the second person’s is the same motion as part of some sort of apotropaic exorcism ritual, they will mean very different things and hence the reasons why both engaged in the exact same motor movements will vary. They will thus have different causes in this sense (which is the sense of cause the social sciences use). Unlike the tree in New Guinea and the tree in New York, the movement of the two people cannot be explained by a neutral, scientific uniform language of mechanical causes. What “caused” one human motion and what “caused” another (in the sense of what they mean per why the person did what they did) cannot simply be explained mechanically but must appeal to the conventions of the community and the network of public agreements within which each event occurred (hence why the two senses of causation in the natural vs. the social sciences are not the same). And, per Wittgenstein’s Private Language Argument, that network of public agreements simply *cannot* be a discreet neural firing. As Dan K. pointed out:

    “Intentional states are inherently linguistic, insofar as they involve our representing states of affairs in various ways, and are thus necessarily rule-governed. For reasons that I am not going to state explicitly here, rule-following is an inherently public activity, from which it follows that one cannot follow a rule (and thus, employ language) from “inside” one’s own mind.”

    Again, from the essay:

    “Actions also are not properly identifiable as discrete objects or events, for they too are as much a matter of interpretation – of how things are represented in the public language – as they are a matter of motor movements. …there must be something else that makes these motor movements count as raising one’s hand, asking a question, requesting permission to go to the bathroom (the latter two of which, notice, can also be effected by one’s hand raising). That something else would seem clearly to be various interpretations that are warranted, given relevant features of the context and the linguistic rules and conventions of the speaking community. But this means that the action of raising your hand is no more a discrete, localizable event in the world, then your desire to ask a question or go to the bathroom is a discrete, localizable event inside your head.”

    Liked by 1 person

  7. Mpboyle56

    It is alluded to, as a reconstruction of what the rule must have been (hence the use of the phrase “I may suppose” by Kripke when referring back to the rule as originally taught). ”

    No, because you have already said that there is no assumption about the language that was associated with the past usage of the rule. So what past usage of the word ‘count’ is being.referred to?

    “This conflates two entirely different things. “,

    No, my statement does not even refer to those things you are talking about. I am saying one specific things – that there are no such things as discrete localisable events

    Like

  8. @Robin Herbert:

    “No, because you have already said that there is no assumption about the language that was associated with the past usage of the rule. So what past usage of the word ‘count’ is being.referred to?”

    I clearly stated that the word count referred to the earlier explanation of the rule and that the word need not be cogitated in subsequent instances of rule-following. Viz.:

    “He does not assume this. “Count” here refers to a word in the earlier explanation of the rule. One need not per se cogitate the word during the activity- this is why Kripke refers to the internalization of the instructions.”

    “From his description it is clear that the explicit rule (which, per Wittgenstein, is a public agreement) was learned in the past (meaning: initiated into that particular public agreement) and that the rule only really comes to mind when one is specifically asked about one’s behavior. Otherwise, as with many other things, we follow the rule/agreement without necessarily thinking of the rule or when we learned it.”

    ____________

    “No, my statement does not even refer to those things you are talking about.”

    That’s right, because it incorrectly assumes that the same model of explanation can be used for both, which it can’t because of the reasons I gave.

    “In physical sciences we can use a simplification which acts like a discrete localisable event and I don’t see why the same thing cannot be done with mental events.”

    The above statement elides the distinct difference between explanations of human behavior and physical events, even in terms of a pragmatic shorthand. Arguing that we can do the same in both cases ignores that both are fundamentally different. In the case of the physical sciences, the shorthand is a stand-in for highly complex lower level physical processes which are nonetheless explicable in scientific terms consistent with the shorthand. In the case of “mental events,” the shorthand comes nowhere close to even approximating what’s happening. What is happening is different in kind altogether and the shorthand is thus completely and utterly misleading.

    Liked by 2 people

  9. mpboyle56

    I clearly stated that the word count referred to the earlier explanation of the rule and that the word need not be cogitated in subsequent instances of rule-following. Viz.:

    “He does not assume this. “Count” here refers to a word in the earlier explanation of the rule. One need not per se cogitate the word during the activity- this is why Kripke refers to the internalization of the instructions.”

    And I pointed out, just as clearly, that said explanation is not “earlier”, it is given in response to the challenge and is therefore couched in present language

    By the ground rules the sceptic is not doubting present language.

    And you have clearly stipulated that there is no assumption about any specific word used with respect to any past usage of the rule, which would include the point at which the rule was internalised.

    So these possibilities have already been dealt with.

    So, again, what is the past usage of the word ‘count’ being referred to that is:

    1. Not a reference to an explanation given in response to the sceptics challenge
    2. Not a reference to any past usage of the rule
    3. Not a reference to the point the rule was internalised.

    That’s right, because it incorrectly assumes that the same model of explanation can be used for both, which it can’t because of the reasons I gave.

    I am not sure how that even refers to what I said. Here it is again “there are no such things as discrete localisable events“.

    Now if you will read the part of Dan’s article that you cut and paste, you are essentially saying that the claim “there are no such things as discrete localisable events” makes the assumption that some things are discrete localisable events.

    Like

  10. Here is what I am saying:

    1. There is no such thing as a discrete localisable event
    2. If there is no such thing as a discrete localisable event then there can be no distinction based on one class of things being discrete localisable events and the other not.
    Conclusion: there can be no distinction based on one class of things being discrete localisable events and the other not.

    I am not sure why that is even controversial. You can’t treat the consequent in 2 as though it were a stand alone claim. In particular, if you claim that the consequent of 2 is false on the basis that the antecedent of 2 is false, then you have simple restated 2 in the contrapositive form and we are in complete agreement.

    If you want to abandon a distinction based on one class of things being discrete localisable events and the other not, and decide upon an entirely new distinction, then you cannot treat my argument as addressing what is an entirely new claim.

    As I understand the new claim, it appears to be based upon “cause” being misleading because you have decided to overloaded the term with assumptions about complex lower level physical processes.

    OK, but that goes for every single word we use. If you mean one thing by a term and I mean another then we will mislead each other. Obviously I can only counter that we can use “cause” in both cases just as long as we don’t overload the term with all those extra meanings that are not strictly implied by it.

    Like

  11. “And you have clearly stipulated that there is no assumption about any specific word used with respect to any past usage of the rule, which would include the point at which the rule was internalised.”

    Here’s what I said:

    “From his (Kripke’s) description it is clear that the explicit rule (which, per Wittgenstein, is a public agreement) was learned in the past (meaning: initiated into that particular public agreement) and that the rule only really comes to mind when one is specifically asked about one’s behavior.”

    “Part of what this involves is the question of justification of the interpretation of rules, which according to Wittgenstein come to an end with public agreements, since there is no deeper level than this.”

    Kripke:

    “The justificatory element of our use of conditionals…is unexplained. If we take into account the fact that the individual is in a community, the picture changes…. When the community accepts a particular conditional…, it accepts its contraposed form: the failure of an individual to come up with the particular responses the community regards as right leads the community to suppose that he is not following the rule. On the other hand, if an individual passes enough tests, the community…accepts him as a rule follower, thus enabling him to engage in certain types of interactions with them that depend on their reliance on his responses.”
    _________

    “I am not sure how that even refers to what I said. Here it is again “there are no such things as discrete localisable events.”

    You asserted that the idea of discrete localizable events is, essentially, a useful façon de parler in the natural sciences and saw no reason why the same could not be done with mental events, irrespective of the fact that you believed in both cases they were “a simplification.” This misses the central point- what the explanations are about is entirely different in kind, and that difference is above and apart from any notions of whether there are or are not localizable events.

    Like

  12. @mpboyle56

    You seem to have forgotten what we were talking about.

    This is how it began. I said:

    “But why does he assume that when I have counted something in the past I needed to think the word ‘count’ in order to count something?”

    You disagreed, you said (my underlining).

    He does not assume this. “Count” here refers to a word in the earlier explanation of the rule. One need not per se cogitate the word during the activity- this is why Kripke refers to the internalization of the instructions.

    My assumption is that you are still standing behind this statement of yours.

    So you have stipulated that there is no assumption that my use of this operation in the past was associated with the word “count”

    Now, as for the “explanation” Kripke puts this in the account beginning “Well, say, to take it in it’s most primitive for, suppose we wish to add x and y …” Are you going to say that this is an accurate recount of a set of words he gave to himself as an instruction in the past? Obviously not. He is describing it in the language of the present.

    Since the sceptic is not, by his ground rules, doubting any present language, this cannot be the past usage of the word he is talking about.

    And you have stipulated, in your words I have quoted above, that there is no assumption that my past usage of the rule was associated with this word.

    It is no use skipping ahead and quoting any conclusions that Kripke has drawn on the basis of this, I am asking a very specific question about a very specific step:

    So, again, what is the past usage of the word ‘count’ being referred to that is:
    1. Not a reference to an explanation given in response to the sceptics challenge
    2. Not a reference to any past usage of the rule
    3. Not a reference to the point the rule was internalised.

    You asserted that the idea of discrete localizable events is, essentially, a useful façon de parler in the natural sciences and saw no reason why the same could not be done with mental events, irrespective of the fact that you believed in both cases they were “a simplification.”

    You are skipping around to different things I said. It makes no sense to talk about a conclusion that I drew from an earlier step, as though it were a stand alone statement.

    Again: here is what I am saying.

    1. There is no such thing as a discrete localisable event
    2. If there is no such thing as a discrete localisable event then there can be no distinction based on one class of things being discrete localisable events and the other not.
    Conclusion: there can be no distinction based on one class of things being discrete localisable events and the other not.

    The conclusion follows from these premises. Which premise do you disagree with?

    This misses the central point- what the explanations are about is entirely different in kind, and that difference is above and apart from any notions of whether there are or are not localizable events

    Now I am confused. So you are saying this distinction has nothing to do with whether or not there are discrete localisable events?

    Here is Dan’s last paragraph:

    Summarizing this point and wrapping up this “sketch,” for α and β to be related as cause and effect, at a minimum, α and β, must be discrete, localizable, events. One trouble with the idea of the intentional causation of action, then, is that neither intentional states nor actions can be properly characterized as discrete, localizable events, and to think that they are is to commit what Gilbert Ryle has called a “category error,” one that is easy to make, for the sorts of reasons that Wittgenstein devoted a good portion of his philosophical work to describing.

    So you are saying that the sense of this does not depend at all on there being discrete localisable events? Right?

    Dan, do you agree?

    If so, I am confused by this last paragraph. Surely if, for for α and β to be related as cause and effect, at a minimum, α and β, must be discrete, localizable, events, then either there is such a thing as discrete localisable events or else there is nothing at all that can be related as cause and effect.

    Like

  13. Robin: Yes, *one* of the arguments that intentional reasons are not causes and actions not effects is that neither constitutes a discrete, localizable event. Of course, that isn’t the only argument, so how much focus one should put on it might be a matter of dispute.

    Like

  14. Robin:

    The word “count” occurs at the beginning (“the earlier explanation of the rule”) but is not necessarily called to mind thereafter per the learning of rule following behavior (the internalization). I don’t think about all the rules for driving when I drive and yet I learned them at some point in the past.

    You assert that physical events are not discrete and so the distinction fails. I am saying (and I’m going beyond what Dan wrote, I think) that even if you are right and they aren’t discrete, physical events (say, some energy field network or whatever) they still aren’t the same and thus your assertion that we can describe both with a useful fiction fails to see that social science explanations would still be fundamentally different.

    Like

  15. Daniel,

    You appear to put a lot of stress on it, it was the whole of your summary.

    I am not disagreeing that such a distinction exists and that it makes a category error to ignore it. One only need read Michael Shermer’s “evolution caused the Mutiny on the Bounty” article http://www.michaelshermer.com/2004/02/bounty-of-science/ to see what nonsense one can talk to have these confused.

    But, as I said, I don’t think there is such a thing as a discrete localisable event and therefore such a thing cannot be the basis of such a distinction, nor for the ruling out of using “cause” in one circumstance or another.

    Like

  16. Also, the claim that “the shorthand is a stand-in for highly complex lower level physical processes which are nonetheless explicable in scientific terms consistent with the shorthand.” implies the hypothesis of the unity of science.

    If our hypothesis is the disunity of science then it cannot be regarded as a stand in for any lower level processes.

    Like

  17. @Mpboyle56

    The word “count” occurs at the beginning (“the earlier explanation of the rule”) but is not necessarily called to mind thereafter per the learning of rule following behavior (the internalization).

    Again, there is no “earlier” explanation of the rule referred to. There is the explanation of the rule, given in response to the challenge and therefore in present language, which is not to be doubted.

    If this refers to the time in the past, when the rule was initially internalised, then the claim that the word ‘count’ was involved is clearly an assumption, as I originally stated.

    Like

  18. One feels that the sceptic would like to say “Oh, come on, stop being so difficult – you know what I mean, even if I can’t express it exactly”. But unless he would extend the same leeway to me, why should I extend it to him?

    My scepticism is towards him – I doubt that he really has a challenge that he can understand if he cannot put it exactly into words.

    I can say to him “Each word I use in my discussion with you right now is used as I understand it in the present. When I speak of my recollection of the past, I do so in the language of the present. Thus, if you are saying that I might have used a word differently in the past, I say ‘What of it?’. I am speaking to you now in the language of the present. Unless you can identify any instance when I have used the word ‘count’ or ‘plus’ or anything else in the past, then you really have no point’.

    So, my response to the first part of the challenge goes:

    “My correct and therefore complete recollection of my prior usage of the operation I currently call ‘plus’ accords exactly with my current understanding of the operation . In particular I did not test the condition which would necessarily have been part if any correct usage of the operation you call ‘quus’.

    By the ground rules the sceptic must agree that I have this correct recollection of my prior usage of the operation.

    So if I had previously used quus then either :

    a. I did not test the condition and used quus wrongly and therefore my mathematical ability is in doubt

    Or

    b. I used quus correctly and did test the condition and forgot about it and therefore my memory is in doubt.

    By the ground rules neither my memory nor my mathematical ability is in doubt, therefore I did not use quus.

    Like

  19. Hi Dan,

    You state that the concept of billions of firing neurons giving rise to human (or animal) behavior is “little more than a version of the genetic fallacy.”

    A fairly standard theory of the genesis of consciousness, human thought and culture is as follows: life evolved out of the matter of the cosmos, based on DNA, competing for resources in order to survive. After billions of years, neurons developed, the most complex of cells. Since then increasing complexity apparently accelerated. Homo is now at the pinnacle of complex social orders, characterized by outrageous abilities to communicate through language, music, art, emotions, fashion etc. This is in fact the glue that keeps the whole thing going for us. Our penchant for disagreement on any complex question is also unique, often with tragic results.

    In spite of mountains of investigational evidence, the details of this story are still unknown. It is generally agreed, however, that the increasing size and complexity of the cerebrum parallels the complexity of behavior. Different parts of the brain have different functions as can be seen on intra-vital study. Damage to specific parts of the brain, with loss of electrical activity, give rise to specific abnormalities in thought, sensation or behavior. The brain is the central processor that integrates inputs with outputs, both internal and external. Most of its activity is not ‘conscious’, i.e. is not identified in conscious thought.

    On the other hand, it is pretty straightforward to realize that all cognitive learning is public or cultural, i.e. learned from others: language, rules, values, mores, style, accent. Even modules of behavior can be copied, in toto, from parents and other influential figures. So sometimes it is correct to say that a decision was totally conditioned by public expectations. However, in each and every case, the actor has the ability to critically evaluate any anticipated behavior and then decide to modify or veto it. Most of the time the critical evaluation is pretty sloppy. This is fairly well documented by psychologists and social scientists.

    Anyway, genetic arguments are not necessarily fallacious. The above scientific arguments seem valid but woefully incomplete. Also, I would be very interested in a summary, if it is available, of Ryle and Wittgenstein’s argument that it is wrong to think that everything really happens inside our bodies. Where else could cultural information processing occur, other than in the brain*? There is a huge amount of interesting but inert cultural information existing outside of our individual bodies waiting to be processed, by one individual at a time. Quite amazing.

    *Advances in AI complicate matters nowadays.

    Like

  20. Liam: The problems for the “inside our heads” view lie in the Private Language and Rule Following arguments, which Ian Ground and I discuss in some detail, in our dialogue on Wittgenstein.

    Like

  21. Hi Dan, as you know I have lived in both camps and so this is of strong interest to me, which makes it frustrating that I didn’t have time to comment.

    I agree with your message in general. I think it should be obvious that there must be differences with regard to “cause” and “explanation” given that most physical sciences study relations/systems of inanimate objects, which move inertially (whether independently or in aggregated group activity), and almost all social sciences study relations/systems of dynamic beings whose only relevant motions (actions) to those fields are goal-oriented or intention-based. The only thing I would want to point out is that not all physical sciences use the same concepts of those either, and would agree with what Per Lundberg said, that “… there isn’t a single, universal definition of “explanation”.”

    Now I will move to areas of disagreement.

    All sciences must choose a window of space, time, and range of subjects for the model they are building, meaning arguably all of them study “discrete, localized things or events.” The question is of magnification: how far one “zooms in” to the phenomena.

    The criticism then cannot simply be that “… neither intentional states nor actions can be properly characterized as discrete, localizable events”. Rather, it is that the phenomena most social scientists are concerned with studying cannot properly characterized at the specific discrete, localized level appropriate for subject matter concerning most physical scientists.

    At some point (even in physical sciences) there can be such limited returns for zooming in, that one is no longer treating the original phenomena under study. For example by the time you are looking at vibrations and rotations in complex molecules you are no longer treating the concerns of macroscopic level thermodynamics. Even in my own work there are places I would be saying to people in other fields of neuroscience…whoooah there! What they study tells me nothing, even if the phenomena must exist deep within the entities I am studying.

    However, none of this bars physical scientists from approaching subjects treated by social scientists, as long as the limitations are expressed properly. For example the neurochemist cannot claim to have found that “love” is a specific (set of) neurochemicals, or the neurophysiologist a specific set of neural functions. But it is not invalid for them to claim they have found physical components which underlie processes that are necessary for humans (assuming a human brain) to manifest or experience that emotion… that such necessary components are restricted to the brain, or all happening “in the head”.

    I want to highly recommend that talk you linked to (and I will like it despite disagreeing with some of what was said). There were a lot of important things discussed. Hacker’s speech in particular (after Tallis) and questions at the end being extremely useful to the debate going on here. I will address it in my next post (maybe later today, but maybe tomorrow), as well as Wittgenstein’s raised arm problem you mention in your essay.

    The distinction Hacker and Tallis made between “necessary” and “sufficient” when discussing brain versus what is a human being or person I think can be used in other places where research interests at one level are not properly addressed by reductive explanations (even within physical sciences).

    Liked by 1 person

  22. it is not invalid for them to claim they have found physical components which underlie processes that are necessary for humans (assuming a human brain) to manifest or experience that emotion… that such necessary components are restricted to the brain, or all happening “in the head”.

    ——————————————————

    I agree with this. If what I’ve said suggests otherwise, then I need to revisit how I phrased things.

    All that I was trying to say with my “discrete, localizable events” point was the following:

    1. Intentional states are not identical with states of the brain.
    2. Actions are not identical with motor movements.
    3. It makes perfect sense to say that states of the brain *caused* — in the efficient sense — various motor movements.
    4. It does not make analogous sense to say that intentional states caused — in the efficient sense — actions.

    The *reason* for this is that just as the mind is not a thing, but rather, “mind” is a name for any number of capacities and dispositions, so beliefs, desires, and actions are not things or events, but rather, complex states of affairs that have both “internal” and “external” components. What makes some set of brain states *count* as a specific belief or desire is that they involve representing things in various ways, representations that involve following rules that are inherently public/social in nature; likewise, what makes certain physical movements count as a specific action, is that they are interpretable in various ways that likewise involve following rules that are inherently public/social. Thus, it makes no sense to speak of the belief as being “in” a person’s brain and it makes no sense to speak of the action as being identical with a certain set of physical movements.

    Like

  23. Hi Dan, in my first reply, I was sort of acting as a grammar or word police, which in all fairness I believe Hacker can also be accused of in some of his criticisms of neuroscientists. The “discrete, localized” terminology you had used was not as useful (from the point of view of a scientist) as the explicit statements you just gave here.

    So I do appreciate your reply to me, which I think makes your own position that much clearer.

    And I should say I thought the video cleared up a few other questions since I assume you are arguing what they argued, which is why I really recommend it. Of course they also started by talking in ways I wasn’t sure about… until the question period where their answers (one it seemed to a neuroscientist) crystalized their meanings for me. It often feels like scientists and philosophers are talking past each other until forced to compare notes about the legends on their maps.

    Two issues remain, which I will address in a larger reply tomorrow. These touch on points 1 and 4 above, as well as your final sentence. Some of it may seem a bit like more word policing, but one I think is a logical issue. The logical issue being with Wittgenstein’s arm problem. You may very well have adequate answers for what I will raise, but right now it remains unclear. Again for the most part I agree with what you are arguing. This is a job of kicking tires and pointing out things that can be tightened.

    I’d write more tonight but the intentions of my gf will be controlling the rest of my actions for the evening 🙂

    Like

  24. These are very subtle arguments that are quite difficult to articulate clearly. Hacker, I think, is particularly good in this regard, which is why I always recommend him, on these topics.

    The basic category errors and fallacies that are involved in the move from this kind of folk language to the technical languages employed by the sciences strike me as very important to identify. Otherwise, it is unclear what, exactly, our sciences are telling us — and what they are not telling us (and cannot tell us) — and this engenders frustration, when we find that even when they are at their most successful, they haven’t told us what we thought they would.

    These problems are the most egregious with those sciences that positions themselves right at the line between the folk and the scientific — Ev Psych, Ev Soc, Cog Sci., and the like. It is almost as egregious in those on the folk side who suffer from science envy, examples of which are legion in the social sciences.

    Like

  25. What makes some set of brain states *count* as a specific belief or desire is that they involve representing things in various ways, representations that involve following rules that are inherently public/social in nature; likewise, what makes certain physical movements count as a specific action, is that they are interpretable in various ways that likewise involve following rules that are inherently public/social.

    http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.135.1244&rep=rep1&type=pdf

    “the honeybee dance language is a formal apparatus which itself is context-free, in such ways that it can, in local instances of its operation, be sensitive to and exhibit its sensitivity to various parameters of social reality in a local context.”

    Searle (“What is Language” 2009) puts his finger on some of the differences between prelinguistic and linguistic intentionality, but following inherently public/socially agreed rules is not one of them. He is also happy to use the terms belief and desire as summaries of the internal states of dumb animals that lead to subsequent purposeful actions.

    I think the causation seen in the bee language collection is just as complicated to talk about as human social causation, but I think it’s useful when we start discussing evolution of bee language, genetics etc/

    Liked by 1 person

  26. Hi Dan, I agree with your assessment, the dangers lie particularly where sciences treat subjects that are at the line (often crossing) between the folk and the scientific. Hacker and Tallis were good, Hacker being better in that particular video clip because he went into more expanded explanations. He certainly revealed some of the errors committed by those in the sciences, with either careless thinking or careless word choice. However he revealed some errors of his own. I think this is where philosophy and the sciences really need to get together to work out a common “legend” so we are not talking at cross purposes, often making false assumptions of what the other is saying, and basically confusing everyone in the public in the process.

    In trying (and failing) to write three concise replies, I think I will skip to assuming the possibility of a semantic/conceptual difference is at the heart of my “issues” and make very brief statements which you can clarify.

    Wittgenstein’s arm. I think a mistake is made when it is said…

    … when I subtract the fact that my arm goes up, from the fact that I raise my arm, and the answer that tempts us is that what is left over is a prior, mental act of intending to raise my arm. We’ve seen, however, that such an answer must be a mistake –that we cannot construe intentional states like “intending,” “wanting,” “hoping,” and the like as discrete, localizable, internal events – and that there must be something else that makes these motor movements count as raising one’s hand, asking a question, requesting permission to go to the bathroom…

    That appears to be arguing if a phenomenon has extensive qualities (or many impinging factors), all of the qualities must be considered in any explanation related to that phenomenon or a mistake is being made. Basically the pesky kid always asking “but why” can keep tripping any attempts to start an investigation, by demanding more than was necessary for the original investigation (though admittedly the damn kid is right there is more).

    To be honest I think this can happen in both directions with physical scientists asking but why to zoom in on the mechanical aspects, and social scientists asking but why to zoom out to find the actual (public) context which the mechanical aspects are facilitating. Both are “real” but need to disregard explanations in the opposite direction to deliver any coherent answer.

    I think it is fair to say that “A complete description/understanding of intentional states or actions will not be identical to brain states or motor movements.” This gets to the “mere” problem Hacker was discussing. However, to simply claim intentional states and actions are not identical with brain states or motor movements, creates the suggestion they have no direct correspondence with them.

    Like

  27. Dan,

    Thanks for the references. The talks of Tallis and Hacker are very interesting, clear and partly persuasive. In essence, the crypto-cartesian error of conflating person, mind or brain is a potential reductionist beast. Agreed.

    Tallis has a feast day pointing out all the unexplained mysteries glossed over by those that pay too much attention to neuroscience. This would only be news to those that are naive about the present state of science. (He floundered a little bit when it came to the ‘finding problem’ – there are, it is now recognized, trillions of ‘clocks’ running in our bodies to ensure that everything is perfectly synchronized.)

    Hacker could not resist in ridiculing those blind neuro-economists, and their ilk, for their limited understanding of reality as he sees it. (He curiously said we do not process information.)

    I respect their concern that scientismists could easily fall into the trap of diminishing the subject of their study: humans as laboratory animals, as it were. However, Hacker and Tallis seem to unwittingly fall into a similar trap: they diminish their ‘opponents’ by focusing mostly on any errors. Everyone makes mistakes. As Nietzsche would say, it’s a matter of perspective.

    My simple rule is: pursue and respect as much information as possible. The only conscious decision making that I am aware of occurs somewhere in my brain, itself completely dependent on and modulated by my body. The input of society is fundamental and absolutely vital. There is nothing in this that Hacker or Tallis disagrees with, as far as I can tell from the video. I suspect a modern Wittgenstein would agree too.

    Like

  28. Liam and dbholmes: the category errors and fundamental fallacies that Hacker uncovers are foundational and not just a matter of a “pesky kid asking ‘why’?” or of merely focusing on errors. They get directly at the kind of enquiry we should be engaged in.

    Indeed, this is at the center of the battle, within the social sciences, between quantitative theorists and qualitative or “narrative” oriented social scientists.

    Liked by 1 person

  29. Hi Dan, my long response just got eaten so I will keep it short. I wasn’t referring to Hacker and Tallis’s discussion with my pesky little kid comment. As I said I agreed with some of their positions, especially Hacker… and yes that includes category errors and some fundamental fallacies. He was right to go after fMRI, for example, but (and this is where he was wrong) seemed to miss that there are plenty in neuro already making the same arguments he gave (including Ramachandran who for some reason Hacker dissed).

    The pesky kid thing was not a slam on anyone, or this kind of discussion (which I think is important). It was an analogy to an effect the raised arm problem seems to create for any scientist (social or physical).

    Like

  30. I’ve been considering the recent phase of this discussion with great interest. I do get the sense that this dispute in academia is actually quite superficial in the end, though very damaging nonetheless. Here are some thoughts to potentially help:

    1.) All sides that can, need to formally acknowledge that they are indeed naturalists. This should especially be true for the followers of Wittgenstein, since they tend to leave us uncertain. From the differing magnifications concept that DB brought up, here I’m talking about infinite magnification. Thus all of reality can be said to reduce, not that any human could ever gain such understandings.

    2.) The neuroscientist’s level of magnification might be called 1000X, and they need to formally acknowledge an inability to reduce things down very much. “Intentionality,” for example, simply does not exist from such a level of magnification.

    3.) In normal existence there is no magnification, and so things can seem quite impossible to understand — here all can be questioned (except “I think”). Though existence will indeed be filtered through our senses, our memories, our languages, and so on, we must not use this as an excuse to stop trying to figure things out. Instead we must take what we think we know, and use this to evaluate ideas that we’re not so sure about.

    4.) Beyond DB’s magnification concerns, I like his “legend for the map” suggestion. This conforms with my theory that there are no “true” definitions, but only more and less “useful” ones. Thus I believe that we must always accept a given speaker’s definitions in the attempt to understand that person’s ideas.

    Like

  31. dbholmes:

    Let’s put this another way. What I take from Hacker — and from my understanding of Wittgenstein — not to mention from my own intuitions and understandings on the subject, is that the investigation of human action is to be conducted primarily at the sociological level. This is because of (a) what we engage in the investigation *for* (in my view, intelligbility rather than causal explanations), and (b) the nature of the “objects” of intvestigation (intentional states and actions).

    In this sort of endeavor, brain science is largely beside the point. It is much more relevant in a different sort of investigation — that directed towards understanding the mechanics of human bodily movement — which I have maintained is not analogous to understanding human action.

    Like

  32. Hi Dan, I basically agree with everything in your reply. Indeed your first two replies to me already tightened things up, or defined the situation being faced quite well.

    My last couple comments have been directed at a potential problem I saw arising in point (b) of your latest reply, in how intentional states and actions got defined (particularly intentional states). The problem does not undercut your overall claim, and does not reject the definition obtained (particularly for the purposes of the social sciences). At most it would call for a slight refining of the statements.

    In short, the method of argument itself (the raising arm problem) seemed problematic in that it appeared to make (as stated) a claim too far, or too strong, which was not necessary to make your point. Since I do not seem to be able to make my case understandable for what problem I feel it causes, I will let it go. This is not to blame you at all for any misunderstanding, the issue is subtle and every time I try to write something that would be more clear it is wayyyyy too long. And in the end I could still be wrong!

    I definitely look forward to your work with Massimo on this.

    Like

  33. Daniel,

    I find your comment just above to DB helpful, and I certainly do want to agree, but I’m not entirely sure what I would be agreeing to given the different ways in which your words could potentially be taken. Yes perhaps “human action” needs to be understood sociologically given lower magnification. And then you did mention that brain science is for “…understanding the mechanics of human bodily movement — which I have maintained is not analogous to understanding human action.” So yes, this is coherent with “movement” on the psychology side, and “action” being neurological. But then how might you emphasize to the other side that you’re leaving your definitions responsible? Could you state, for example, that from an “infinite magnification,” all of reality would then fully reduce?

    I also worry about potential Wittgensteinian anthropocentrism. For example, I have a theory that there is only one essential means by which the conscious entity figures things out — it takes what it thinks it knows, and then uses this to assess ideas it’s not so sure about. But my perception is that Wittgenstein would deny human connections with the rest of nature, simply given that we’ve developed languages, cultures, and so on. If there are such blanket denials (which the neurologist most certainly would not deny) then yes, I must object in this regard. Surely the stuff between the ears of the conscious entity, will indeed be what runs this particular kind of machine, and thus there should be important functional similarities between us all.

    Like

  34. Eric: No, “action” is not neurological. Action consists of motor movement interpreted in certain ways. These interpretations are *constituitive* of action and thus, cannot be reduced to mere motor movement.

    Like

  35. Aha! This is the crux of our different understanding: that brains do not process information is completely anathema to my understanding.

    Your approach starts from the socio-cultural end and works down. All information is semantic.

    I would suspect that every physicist views information processing to occur at all levels, from the most microscopic to ultimate macroscopic.

    As a biologist I see all the levels and types of information intersecting at the human level. We cannot afford to ignore any sources of information if we are interested in a full understanding of ourselves and our world.

    Like

  36. Nothing counts as information until it is interpreted. There is nothing inherently representational and thus, nothing that inherently counts as information.

    What makes a string of o’s and 1’s “information”? That people interepret it in particular ways. What makes a number of electrical impulses count as “information”? That people interpret it in particular ways.

    The idea that some substance or process could be *inherently* informational, without any interpretation strikes me as nothing but sheer stipulation — and incoherent too.

    Liked by 2 people

  37. “Language sets everyone the same traps; it is an immense network of easily accessible wrong turnings. And so we watch one man after another walking down the same paths and we know in advance where he will branch off, where walk straight on without noticing the side turning, etc. etc. What I have to do then is erect signposts at all the junctions where there are wrong turnings so as to help people past the danger points.”

    With those wonderfully evocative words your hero describes our difficult social reality. The task he sets himself seems to me to be impossible, unfortunately. There are some fundamental difficulties that must first be resolved. 🙂

    Like

  38. “The idea that some substance or process could be *inherently* informational, without any interpretation strikes me as nothing but sheer stipulation — and incoherent too.”

    This makes me wonder if such a view isn’t ultimately grounded in a view of nature as a book to be read, a metaphor famously used by Galileo, but whose origin is ultimately medieval.

    Like

  39. Daniel,

    Interesting and I enjoyed reading.

    I liked a lot of the comments, nothing to add except that for me the word ‘intention’ is especially problematic.

    I look forward to more on the subject.

    Like