by Daniel A. Kaufman
___
More and more it seems to me that the way in which philosophy has gone the most wrong is in trying to assimilate human action with the motions of bodies, as understood in the natural sciences. This effort has made a mess of the philosophy of mind, the philosophy of action, as well as ethics, and has also led scientists, especially working in the so-called “cognitive sciences,” down any number of empty rabbit holes. It is the product of a more general mistake that I have been talking about recently and that is the effort to reduce the Manifest to the Scientific Image – either wholesale or piecemeal – or simply eliminate it in, leaving the Scientific Image as our sole picture of the world, the motivation for which seems to be a somewhat inchoate and entirely unargued for scientism. One sees this in everything from the efforts to “naturalize” philosophical disciplines like epistemology and ethics to the insistence on monistic, materialist ontologies, which means that one must either reduce or in some way fictionalize anything that cannot be characterized as being “made of” matter or energy and explained in the manner characteristic of the natural sciences.
The trouble, of course, is that the world includes selves and the social and cultural realities that selves create, and neither selves nor social/cultural realities can be rendered or explained within the Scientific Image. The resisting element, it seems to me, is representation and the role that it plays with respect to people’s actions and their reasons for them.
Representations
To represent a thing or a state of affairs is to consider it from a particular point of view and under a particular description that depends upon that point of view. One of the things that is notable about this is that representations of things are neither generalizable nor transferable. That I have a certain belief about something, under a certain representation, does not mean that I have the same belief about it under a different one. After watching Madonna’s speech at the Woman’s March that took place in the wake of Trump’s election, for example, I might believe that she is a self-important bore, but it is entirely conceivable that I might not think this of the Material Girl, if I didn’t know that they were the same person. And what is true of beliefs is also true of desires. It may be true that I want something, when represented a certain way, but would not want the very same thing, represented differently. George Orwell, in recounting his time spent fighting in the Spanish civil war, said that he was quite in the mood to shoot “fascists,” but when he saw a man from the opposing side emerge from his trench, pants halfway around his ankles, he found that he couldn’t think of him as a “fascist” and that he was disinclined to shoot him. (1)
In the philosophy of language, the realization that we cannot freely exchange different representations of the same thing is described as a kind of “opacity” that is created in “intensional” – by which is meant representational – contexts. The reason I am highlighting it here, is because W.V. O. Quine, the father of contemporary philosophical naturalism, maintained that this way of characterizing the world is incompatible with natural science, in which explanations are not and should not be dependent upon any particular representation of a thing or an event and thus, on any particular point of view. Hence his effort to purge intensional idioms from the language of science and his insistence on a purely “extensional” semantics, by which is meant one in which every term is substitutable for every other co-referential term, without changing in the truth value of the sentence in which the term appears.
I agree with Quine that the language of natural science is and must be extensional, but draw a very different conclusion from this than he does. Because the language of people and their actions and reasons cannot but be intensional, and because I don’t think that the world of people, their actions, and the social realities they create by way of those actions can be conceived of or treated as “fictions” – indeed, in my view the very suggestion is absurd – natural science is always going to be the wrong framework within which to understand them.
Reasons
What makes X a reason for doing something is that X is represented as a means to some other thing, Y, that is represented as an end. Doing things for reasons, then, is always teleological, which means that explanations in terms of reasons are always going to be teleological in nature. If I assault someone, it’s because I represent him as a menace, and if I aid someone, it’s because I represent him as deserving of assistance. I do both because I represent the relevant states of affairs – where menaces are removed and those deserving assistance aided – as goods.
This alone should make us realize not just that accounting for actions in terms of reasons is nothing like the explanations that we give of physical motion in terms of causes, but that the role reasons play with respect to actions cannot be causal, at least not in anything like the sense meant in the natural sciences, which have eschewed final causes for centuries, biology notwithstanding. Explanations in biology may be teleonomic, but they do not involve representation and consequently, are not intentional.
Also unlike causes in natural science, the primary epistemic role of reasons is not to predict future actions, but to render them intelligible from the actor’s point of view. When we ask a person for his reasons for doing something, we want to know why it made sense to him. Our interest is not in the action itself, but in his performing it. Whereas the natural sciences provide us with myriad events that are sufficient for subsequent ones and thus of great use in making predictions, our accounts of actions in terms of reasons mainly provide us with intelligible narratives, in which we and others feature, as if we were characters in a story.
Actions
An action is an event represented in a certain way and performed by a person for certain reasons. The very same motor movements may constitute entirely different actions, depending upon the relevant representations (including representations of the relevant contexts), which is sufficient to show that actions are not reducible to or identical with mere events. For example, to raise one’s hand is to ask a question, given one set of representations, but given a different set, to raise a hand could be to bid on an item at an auction or to request permission to go to the bathroom.
Actions, then, cannot be assimilated to the “motions” that it is the task of the natural sciences to explain. Rather, they are one part of the intelligible narratives I just mentioned, which also include selves and their reasons for acting. We want to render a person’s actions intelligible from his point of view, in good part, because we want to render both his actions and his point of view intelligible in a larger human story that is valued in good part for its meaningfulness. The understanding sought, then, is much more like that involved when we read literature than of the kind involved when we are engaged in scientific investigation, and this, in my view, is what fundamentally separates the natural sciences from the social sciences and humanities. The former are primarily about the identification of causes, while the latter are primarily about the development of meaningful stories about ourselves and the natural, social, and cultural worlds that we inhabit and which make up “the world,” as conceived in the Manifest Image.
From the 1960’s on, this narrational view of persons, their actions, and the stories into which they figure has been viewed with suspicion as a kind of “epiphenomenalism,” at best, and at worst as nothing more than fiction, obscuring the physical-mechanical “reality” that lay underneath. (2) And it is the supposed intolerableness of this that has led virtually all of analytic philosophy to flock to Donald Davidson and his argument in “Actions, Reasons, and Causes” (1963) that reasons should be understood as causes, in the scientific sense. (3)
It is as a force in opposition to this view that I increasingly see my philosophical work, and especially, my recent emphasis on the reality of agents and actions and of the human narratives in which they figure.
Notes
- George Orwell, “Looking Back on the Spanish War.” (1943)
http://orwell.ru/library/essays/Spanish_War/english/esw_1
- Giuseppina D’Oro and Constantine Sandis, “From Anti-Causalism to Caualism and Back: A History of the Reasons/Causes Debate,” in their Reasons and Causes: Causalism and Anti-Causalism in the Philosophy of Action (Palgrave Macmillan, 2013), p. 7.
- https://bibliotecamathom.files.wordpress.com/2012/09/actions-reasons-and-causes.pdf
Comments
50 responses to “Representations, Reasons, and Actions”
“George Orwell, in recounting his time spent fighting in the Spanish civil war, said that he was quite in the mood to shoot “fascists,” but when he saw a man from the opposing side emerge from his trench, pants halfway around his ankles, he found that he couldn’t think of him as a “fascist” and that he was disinclined to shoot him. (1)”
Newtons 3rd Law, if real, precedes Newton: “For every action, there is an equal and opposite reaction.”.
If Time is no obstacle in the human tragic drama, Orwell pulling the trigger requires that his victim return and take revenge in the next iteration, ad infinitum, endless loop.
His disinclination to shoot the fascist, if acted upon, stops the endless loop of revenge. Perhaps the killers motivation is love, forgiveness, pragmatism all.
Humans engage in tragic and comedic drama mostly, not counting the cost, not drawing an if-then-else philosophic virtue decision process map to guide their behavior before killing.
Human imagination in film frequently points to this undoing of remorse and regret. And modern computers have created the “undo” function.
Life is more about human drama than number.
This is fascinating stuff.
Coming from anthropology, I’m surprised about Quine’s conclusions- I would question (and this is a genuine question!) whether the hard sciences would even care about the representation of, for example, the raising of a hand beyond the motion of the joints and the flexion of the muscles – in other words, that the hand *can* rise. On the humanities side, Geertz wrote about representation when he talked about “the wink” and the many possible things it represents when he was putting forth his ideas about thick description.
For anthropology, obviously, this remains a sticking point. We all make a stab at thick description, but with the understanding that we can never truly understand the actions/understandings of representations of others – we can only try. Whether we can even understand our own actions is up for grabs…
Although now that I think about it, I’d say that I understand my own actions differently in time – my understanding of my actions today is different for current-me than it will be for future-me.
You touch on some excellent points here. Science deals with the low hanging fruit and the hard stuff that is left is still the province of philosophy. Wittgenstein, in “Philosophical Investigations”, dealt with these issues, especially in his private language argument. Human actions are rule -governed. We can only understand what it means to follow a rule, by situating the rule following action within a wider community. To quote Kripke on Wittgenstein: “Wittgenstein’s skeptical solution…depends on agreement, and on checkability – on one person’s ability to test whether another uses the term as he does. The solution turns on the idea that each person who claims to be following a rule can be checked by others.
So this is an attack on the idea that psychology is a science? Freud thought he was developing a science, but that it would be a 100 years before his “top-down” science of the mind would join up to “bottom-up” neurology. His main point that influenced an entire century was that the real causes of both individual behaviour and culture are not reasons of the kind you are talking about.
I reject the idea that social sciences — of which psychology is one — are part of or like the natural sciences, yes.
All sorts of things that are rubbish influenced all sorts of centuries. Not sure what that’s supposed to show.
My views in this area are very much influenced by Winch’s The Idea of a Social Science and its Relation to Philosophy (1958), among others, including, notably, Collingwood.
I’m in fullest agreement here. I’ve been impressed with the manifest/scientific differentiation of Sellars you have come back to. I feel C P Snow, though wrong in his final estimation, was onto something in his “two cultures” idea and seems to me to relate to this and is not at all dated
“So this is an attack on the idea that psychology is a science? ”
“I reject the idea that social sciences — of which psychology is one — are part of or like the natural sciences”
A useful way to think of it are that there are two broad categories of science, descriptive science and explanatory science. Descriptive science will observe, count, catalogue and categorise. Explanatory science will observe, test, measure and derive causal explanatory mechanisms. The two overlap in the sense that some areas of descriptive science hold out a realistic promise that they will eventually yield to causal explanation.
But we utterly lack any means of deriving causal explanatory mechanisms for the operation of the mind(as opposed to the brain), even in principle, because there is no imaginable[!] way in which we can observe, test and measure the mind, so any study of the operation of the mind can not be an explanatory science. It is at the very most a descriptive science and even calling it that is a stretch, making it more of an honorary appellation.
Let me use an example from the world of computers to illustrate what I mean. I and a team of programmers have created a large suite of programs that control CNC milling machines. Given suitable CAD drawings from Catia(a large CAD system), it can mill the complex tooling used for making press parts in automotive manufacture.
Imagine for a moment all you could ever observe were the complex motions of the milling machines(human behaviour). This is a most impressive sight and well worth watching. It soon becomes evident that behind the motions is intelligent intent. We observe and categorise the intricate paths followed by the cutting heads and note the beautiful products it produces. But how do we explain it? We observe that milling machines often do certain things and produce certain patterns. We describe this operation in clever papers(psychology or social science).
Soon we conclude that is not enough and so we get clever and put probes on the milling machines, detailing its circuitry and measuring electrical currents. We scan it with infrared machines and say, aha, a certain inegrated circuit gets hot when is performs a certain function. And so milling-science(neuroscience) thinks it is explaining the operation of my milling machines.
But I, the programmer know better. To explain the operation of my milling machines you would need to know that somewhere there is a large computer running my complex suite of programs. It produces executable modules that are sent down to the milling machine and controls the milling machine. You would need to know what programming languages were used and how they were compiled. You would need to understand the programming language. You would need to understand the operating system. You would need to read my source code with all my explanatory comments. When you did this you would begin to understand the structure of my programs. You would see the family of classes that I created and how they were instantiated as functional objects. You would see how they all fitted together to perform larger functions. You could then produce a causal explanation for the operation of my milling machines down there in the large tool-room.
But you, the milling machine engineer(natural scientist), have none of this knowledge. And for as long as you observe the operation of the milling machines you have no chance whatsoever of deriving causal explanations. That is because you are looking for the wrong thing in the wrong place. None of the knowledge that would explain the operation of my milling machines is available to you and it will never be available to you for as long as you look at the milling machines. But the milling machine engineer only knows how to look at milling machines. The large explanatory layer of knowledge necessary for explaining the operation of my milling machines cannot, even in principle, be derived from examination of the milling machines.
Dear Labnut, constraining the amount of information available obviously limits the inferences an observer can make in any setting. If a scientist sets out to predict the trajectory of either a fly or a football, their model must incorporate intention-of-action and intention-in-action. I regard biology as one of the natural sciences, even though it may be “special”, so considerations of purpose and function are obligatory.
Dear DanK. I brought up the psychodynamic model because it is a scientific model. You are probably aware of people like Mark Solms, Antonio Damasio, and Jaak Panksepp who are currently providing a neuroscience basis for traditional Freudian concepts such as affect, id, repression, transference etc. What do I mean that it is scientific? All the usual stuff – explaining known facts better than other explanations, making testable predictions, cohering with other bodies of scientific knowledge. So it may not be correct, but it throws up lots of experiments or novel observations we can make. For example if “…the unconscious mind is just that: it’s a mind. It’s not just something automatic, something out of consciousness, something that reflexively happens. It’s an intentional agent that’s part of you, that’s making decisions, driving your volitional activities without your awareness. Most importantly, which is, I think, uniquely psychoanalytical, it resists self-knowledge. It’s not just that it’s outside of awareness, you don’t want to become aware of it.” [Solms in Poulsen et al 2017], then it makes very different predictions than the more usual “automatic processing” type model of much cognitive neuroscience.
http://onlinelibrary.wiley.com/doi/10.1111/nyas.13403/full
Specifically on representation – this can be tested by a wide variety of rigorous scientific approaches. The experiments that psycho- and neuro- linguists do to address the Sapir-Whorf hypothesis are very familiar to me (ie a scientist from outside that field) in terms of design and analysis eg effects of grammar of one’s mother tongue on performance in visual (non-verbal) tasks, or comparing “theory of mind” test performance in deaf children of signing and non-signing parents.
The manifest image and the scientific image appear to me to correspond to a form of epistemological dualism: there is clearly a difference between the world as it is subjectively perceived and the world as it is. The former is almost entirely conditioned by the observer’s cultural orientation. The latter is formulated on the basis scientific information and often contradicts our manifest cultural and social understandings.
There is, of course, a very dynamic interaction between these two images, but in the end the cultural, manifest image is adjusted to fit the scientific image. This is a very slow and arduous process that is not based on much logic or reason.
Our present time is marked by a huge flood of new scientific information in many different areas. This is overwhelming and is creating much stress and confusion in society.
There is, of course, a very dynamic interaction between these two images, but in the end the cultural, manifest image is adjusted to fit the scientific image. This is a very slow and arduous process that is not based on much logic or reason.
= = =
This is exactly what both Sellars and I reject. Hence his “stereoscopic” vision of the two Images.
I’m afraid that there can’t be mentalistic/reductive accounts of representation for the reasons I have laid out — as well as for Wittgensteinian reasons. There are irreducibly social dimensions to representation. That’s the whole point of the private language argument — representation in language cannot be reduced to representations in the mind.
David,
“constraining the amount of information available obviously limits the inferences an observer can make”
It is not constrained in this case, it is entirely missing.
The theories about concept acquisition whatever they might be are stories told by adults for adults on the basis of a wholly achieved conceptual schema. The work of Piaget demonstrates that it takes years to arrive at what we take for granted. The movement is from a world of ‘blooming, buzzing confusion’ (William James), pure qualia perhaps, where there is no strict division between self and not-self; all the way to the outline of a mature mind. This period is one of great plasticity in the brain and it seems to be the case that if a primary language is not learned before the age of ten then it is very difficult to so thereafter.
“An inner process stands in need of outward criteria” (Wittgenstein) but does it do so to the extent that it does not exist until it is named ‘publicly’. If not a ‘beetle in the box’ there is something there. The stages of concept formation in the child suggests that there is always something there.
” We want to render a person’s actions intelligible from his point of view, in good part, because we want to render both his actions and his point of view intelligible in a larger human story that is valued in good part for its meaningfulness” seems more likely we want to know what to do next and how to manipulate the situation to our advantage, see:
https://soundcloud.com/edgefoundationinc/dan-sperber
also in reference to yer recent exchange with Crispin I don’t think the Law exists and that you can know this because you got a speeding ticket as much as you and the cop acted As-If there was such a thing as a law just as we might act as if there is a ghost (holy or otherwise) at work in some situation.
You don’t need the source code of a system to understand it.
I have, more than once, recereated systems where the source code has long gone missing and reverse compiling is not an option.
In that case you have to rewrite the program based on the observed behaviour of the system alone.
I may not use the same class structure etc that the original programmer used, but if I end up with an identically behaving system then what did I not understand about it?
For me this makes it a poor analogy.
Dan,
Breezed through Davidson. will probably need to get back to it. But hoping to read Sellars tomorrow.
Reading Davidson, I confess I find it troubling to find him seeming to collapse motivation into causation. Causation is so cut and dried; motivation is rich, complex, occasionally internally conflictive. Also, although many motivations are psychologically or socially determined, it’s seems weird to deny that some are either self-originating, or responsive to new and unexpected situations. * So it seems odd to find Davidson making what seems to me such an obvious mistake. Your further thoughts on the matter?
In Davidson’s defense, I recognize that he did catch the important point that what we are after is explanations – and particularly explanations of explanation. But I don’t think he recognizes the breadth of that problem, as we see in his essentially univocal reading of “because” – which is in fact a highly ambiguous term in common usage, even though we all know what it means when we use it. (Its usage is highly context-dependent; whereas, on first reading, Davidson’s understanding of it seems context-impoverished.)
-_-_-
* The problem with motivations either self-originating, or responsive to new and unexpected situations is that their predictability decreases considerably; also they often increase the inner conflicts of motivations, leading to unexpected results.
Dan,
On a more general note: I confess that after three or four years reading and discussing these issues, I’m disappointed and annoyed that the scientistic hope for reduction of social issues to psychological to neurological to genetic to chemical to physical – I just don’t get it; it’s so obvious that we’re talking two separate realms of knowledge that the confusion between the two seems exactly that, a confusion.
Yes the manifest image adapts to the scientific – on scientific matters; so we are almost entirely heliocentric as concerns our solar system. But what has that to do with courts of law? With the practice of marriage? With how I interact with my neighbors? Will neuroscience explain that? Even if it could, would it *improve* matters? According to what – purely socially determined – standards?
You know what? I think I’ll drink a beer and go to bed. And I don’t really care what explains that – especially not anything having to do with neurons in my brain (and I certainly don’t care for any medical/psychiatric sermons admonishing me not to).
This ‘I don’t care’ is “the wrench in the human machine, the flowers in the dust bin” – Punk rock remains the cultural proof why scientistic puritans will never succeed in recreating social reality to their prescriptions.
And if they can’t do that, what is their point in uttering normative injunctions disguised as scientific explanation?
My point being that the scientistic reductionism debate actually distracts us from far richer conversations we could have about the human experience, both social and individual.
(Finally: Given the time and effort you’ve put into this, do I detect a major paper or even book in the works? Would love to read it!)
Hi Dan:
As someone brought up, philosophically, on Collingwood and Winch, I think all your points are good ones. Here are two examples that I find useful illustrations of how we have to think when we are trying to understand some social phenomenon.
The first is the discussion of Hadrian’s Wall in Collingwood’s “Autobiography”:
“the many archaeologists who had worked at the Roman Wall between Tyne and Solway had never, I found, seriously asked themselves what it was for. Vaguely, you could of course call it a frontier defence, and say that it was meant to keep out the tribes beyond it. But that will no more satisfy the historian than it will satisfy an engineer if you tell him that a marine engine is to drive a ship. How did it work? Was it meant to work, for example, like a town-wall, from the top of which defenders repelled attacks? Several obvious features about it made it quite impossible that any Roman soldier ever meant to use it that way. No one seemed to have noticed this before, and when I pointed it out in 1921 every one who was interested in the subject admitted that it was so, and my counter-suggestion that it was an ‘elevated sentry-walk’ was generally accepted.”
The details of the example don’t matter, but the method of inquiry — what he called “question and answer” — is what he was illustrating.
The second is from Julius Kovesi’s “Moral Notions”. (Kovesi was my teacher.)
“Suppose someone is trying to buy flowers, but cannot find any. If he then buys a packet of paper streamers we can say that he was trying to buy decorations, but if he comes home with a box of chocolates, we can say that he was looking for a present. If we want to find out what someone is doing who is, say, leaning against a door frame, we need to find out what he would do instead which would amount to the same thing. If he sits down, then we can say that he was resting; if he stands a beam against the frame then we can say that he was supporting it.”
Understanding actions involves understanding what would count as “the same”, and often that can’t be known simply from the observable features of an act. It involves a counterfactual.
Social inquiry is a rational process, but (as Winch argued) its goal is nothing like the universal generalisations that are central to the natural sciences. In Collingwood’s words, “there are no mere ‘events’ in history: what is miscalled an ‘event’ is really an action, and expresses some thought (intention, purpose); the historian’s business is therefore to identify that thought”.
Alan
Robin
“You don’t need the source code of a system to understand it.”
That is rather like saying you don’t need access to my thinking to understand me. Or, all that you need do is observe my behaviour to understand me.
A description of the functions performed by a person or a system is not to understand it but merely to describe it. There is a rather large difference between understanding and description. Certainly the description is part of the understanding but it is only a part.
Because the language of people and their actions and reasons cannot but be intensional, and because I don’t think that the world of people, their actions, and the social realities they create by way of those actions can be conceived of or treated as “fictions” – indeed, in my view the very suggestion is absurd – natural science is always going to be the wrong framework within which to understand them.
——————————–
I guess that you are trying to establish or defend an intentional safe space in which our personal, social and cultural narratives can be fully developed and appreciated. This is a laudable goal since there are constant forces working in the other direction.
The difficulty, of course, is that it is virtually impossible to create any narrative without relying on the use of extensional language. I am not even sure that one can make a wholly intentional sentence in which there is a complete absence of an extensional component: ” I sense harmony between truth and beauty?”
I agree that it is a good fight to keep the excesses of narrow ‘scientism’ at bay. By the same token, however, opposing the restrictions of all excessively rigid philosophical and political interpretations would be equally important. .
Robin,
“You don’t need the source code of a system to understand it.”
Let me give you a concrete example from something I am working on right now. I am working on a GPS tracker that will record my track while doing trail runs. Among other things I want
1) an accurate distance which is not amplified by jitter in the GPS readings;
2) an accurate measure of height gained or lost. GPS is notoriously variable in the way it reports altitude;
3) a good estimate of the Calories burned during my run. All systems I have encountered fail to factor in the height gained or lost, an important consideration in mountainous trail running.
I must apply a Kalman filter to (1) and (2) but with different constants.
I must use one of the Calorie calculation algorithms for (3). This is a tricky problem in its own right.
When you examine my system you will deduce that I have applied some filtering but you will have no idea what kind of filters I use, why I chose them, what filtering constants I used and why I chose them. You will see that I report Calories but once again you will have no idea which algorithm I used, what constants I applied and why I chose them.
In short you will have only a most superficial understanding of my system, so superficial that it hardly counts as understanding.
There is an entire layer of knowledge missing that prevents your understanding of my system and it will never be available to you. Contrary to what you say, this is a good analogy for the distinction between mind and brain.
Robin, I said
“I must apply a Kalman filter to (1) and (2) but with different constants.”
But why a Kalman filter? There are other very good and complex filters I could use. Now I must factor in consideration of the processing power used and how it affects battery life on my smart phone and so there must inevitably be a compromise. This is in turn affected by knowledge of the length of the trail runs I do (> 5 hours) But which compromise did I choose and why?
None of this knowledge could conceivably be available to you and so any claim that you understood my system would be farcical.
You are right in part about my motivations, although that isn’t all there is to it. But I don’t quite understand your criticism, as I never suggested getting rid of extensional language which, as you indicate, would be impossible. My point goes in the other direction: naturalists like Quine want to get rid of intensional language, which I think should be viewed as equally impossible, but recognizing this requires one to take seriously the existence of selves, reasons, and actions.
Thank you! These are excellent!
Yes, all of these “bits and pieces” I’m doing represent an effort to accumulate my thoughts for the purpose of producing a book on the subject
It’s funny, but I blame Davidson much less than the hordes of much less thoughtful people who just raced after him. For the most part, as you know, I am very much on board with much of Davidson’s thought and believe him to be in the top three or four best post ww2 philosophers.
Prof. Kaufman, I sympathize deeply with your ideas.
I would like to draw attention to a way of representing intentional behavior that doesn’t require our representing the agent’s reasons. I don’t think you explicitly mention it, but I think it fits with, and might even enrich, your account.
We can competently recognize intentional behavior without representing the agent’s intentions; we can represent an agent as acting for a reason without representing the agent as acting for THE reason that . . . (where the ellipsis elides a representation of the specific reason for acting). Indeed, you might think that representing an event as an action (as done for some reason or other) opens up the possibility of representing the event as the action it is (as done for THIS specific reason).
(Side note: I think this is one important thing, of several, that Kant was trying to capture in the third Critique with the phrase “purposiveness without a purpose” or “meaningfulness without a meaning.”)
One reason it’s important to draw attention to this way of representing agential behavior is that even though it’s a rich and complicated competence, it isn’t as conceptually and epistemically demanding as representing an action together with its specific reasons is. So objections along the lines of “Your account and other nonreductionist ones like it require all of us, children included, to be mind readers with unrealistically sophisticated or fine-grained conceptual repertoires” won’t get much traction.
Thoughts?
Perhaps a stupid question, but I’m going to ask it anyhow.
I know and have talked with quite a few sociologists and economists. None of them thinks his (or her) work is comparable to the hard sciences. One economist (who has tenure at the local university, teaches macro-economy and is good mathematician) empathically denies his research is “hard science” in the sense of physics or chemistry or biology. If I ask them if the phenomena they study are reducible to physics, chemistry or biology, they shrug and answer that the question is entirely irrelevant for their work. Their work is about something fundamentally different. The sociologist (who has tenure too, who specializes in the sociology of culture but made his phd in the sociology of religion) simply smiled at my stupidity and said: “just try to find a solid, hard, scientific definition of ‘being religious’”.
Does there really exist an honest, viable attempt to reduce “the social sciences” to physics, biology, evolution theory etc.?
My personal (and very anecdotic) experience suggests that it’s a fringe activity in the “what if?” mode.
I don’t have psychologists among my acquaintances, though.
I suspect they don’t for the most part, although people like Skinner certainly did. It’s primarily a philosophers’ fixation.
Seems to me that if we don’t know the other’s reason, we don’t know why he acted.
Hi Couvent2104. “an honest, viable attempt to reduce ‘the social sciences’ to physics, biology, evolution theory etc?”
Well I don’t know if reduction is the right word, certainly there is no reduction in the complexity of the models that would arise ;). I think most scientists would believe in some rough kind of “levels” model, so the interest is in the way the upper levels arise from the lower levels, and how levels interact with each other. But we already know that even low level physics can give (computationally) undecidable processes. If the Cobham-Edwards thesis is right, there is a lot of stuff we will never really know completely. However, I have worked in the area of behaviour genetics – I don’t think the behaviour bit is any less scientific that the genetics bit.
Regarding Alan’s comment on Hadrian’s Wall, how is this different from Darwin looking at a finch’s beak? The idea that scientists put a question to Nature is a common one. It is the same for analysing behaviour of other animals – the reasons for some of them are completely opaque to us, but we do believe that we know the processes that they arose by.
And Labnut, “None of this knowledge could conceivably be available to you” – of course it can be! We slowly chip away at much more complex biological systems. Our models are successively better approximations – but with the caveat we may only ever approach to a certain level of error.
Couvent, Dan-K
David Chalmer’s survey(https://philpapers.org/archive/BOUWDP) gives a broad idea of what philosophers think. To the best of my knowledge there is no equivalent survey among natural scientists, so it is hard to say what they think.
From Chalmer’s survey, Question 16:
“16. Mind: physicalism 56.5%; non-physicalism 27.1%; other 16.4%.”
The illuminating question is this one
“26. Teletransporter: survival 36.2%; death 31.1%; other 32.7%.”
Philosophers don’t agree about much.
David,
“None of this knowledge could conceivably be available to you” – of course it can be! We slowly chip away at much more complex biological systems.”
You have made an emphatic statement of belief and hope, but that is not the same thing as an argument.
My arguments is this:
1) I can find objects in the world(my GPS tracking program) that cannot be properly explained by science.
2) Therefore there exists a class of things that cannot be properly explained by science.
3) The mind likely belongs to this class of things and therefore it is likely that it cannot be properly explained by science.
You contradict my assertion (1) without providing any reasons beyond a hope. That is not a satisfactory reply to my argument. To drive my point home I will expand on my example. When you, the scientist, examine the output of my tracking program(remember that is all that is available to you), you will find mysterious gaps in the track. You may surmise that these were points where I briefly halted my run. But why?
Well, I might have paused to cross a busy road or, I might have paused to empty the stones from my running shoes, a hazard of trail running. But you, the scientist have no way of knowing this.
But it gets worse(for you, that is). How does my program recognise pauses in my run? By examining the GPS recording? Bad idea, because GPS jitter can look like a real pause. So I get clever. I use the acellerometer data from my smart phone to determine whether I am walking, running or standing. But I have in my pocket a Bluetooth button which I can press to signal a pause. Just to make it more complicated, there is also an accelerometer in the GPS button and right now I am debating whether it is better to use this data because the acellerometer in the smart phone uses too much power.
None of these choices that I make is apparent from the output of the program and therefore science can never determine which choices I(the programmer) made and why I made them. Therefore there exists a class of things that the scientific method cannot properly examine and produce an adequate explanation.
Now this is quite a shocking state of affairs for a Scientologist(my name for a scientismist). They are ideologically committed to a blind belief that everything, bar none, is explainable by science. And yet, it is easy, as I have shown, to produce concrete counter examples that science cannot explain. When confronted with these examples I hear them stutter, hem and haw, equivocate and finally retreat into broad statements of wish fulfilment.
“Seems to me that if we don’t know the other’s reason, we don’t know why he acted.”
Right. But I’m not talking about knowing why he acted. I’m talking about something prior: knowing that he acted. I’m talking about our ability to distinguish events that are actions from events that are not. Only if we represent an event as an action does it make sense to ask about the reasons for the action. I was pointing out that we don’t have to know the reasons why someone acts in order to represent him as acting.
And I thought I was offering further considerations in support of your view, or at least bringing something implicit in it out into the open. Here’s what I mean.
The picture according to which reasons are (like) causes is one according to which actions are distinguished from other events by the former’s having special mental antecedents: like other events (the story goes), actions are bare movements, but unlike other events, actions are caused by special things. On this picture, to represent something as an action is to represent it as a bare-movement-plus-a-special-antecedent.
I thought that, on one way of putting your view, we don’t represent actions as bare-movement-plus-something-special because we don’t represent actions as bare movements at all — whether in conjunction with special antecedents or not.
Someone might then wonder: when we represent an action, what are we representing if not bare movement (plus something else)? And that same someone might also wonder: aren’t your ‘reasons’ just the sort of extra things that we add to a bare movement in order to represent that movement as an action?
If we point out that we are able to represent someone as acting without representing his specific reasons for acting, we go some way toward defusing these questions. To represent an event as not-just-bare-movement, we need not represent special extra items, be they causal antecedents or specific reasons. (Not that YOU think of reasons as extra items — I’m speaking the objector’s language here.)
David,
Why do I call scientism Scientology? Scientology(of L. Ron Hubbard fame) have their infamous E-meter, which they used to investigate the thinking of their converts. It has been shown again and again to be nothing but flimflam but it possessed considerable power to intimidate since the analyst could impose on it all his wilful preconceptions. Today neuroscience has rather more advanced scanners. But when it comes to thought it is nothing more than an E-meter in very expensive drag. Some sectors of science have adopted the methodology of Scientology and it is only right and proper that they should therefore be labelled as Scientologists. L. Ron Hubbard would be proud of today’s neuroscience and he can rightly be regarded as the founder of neuroscience 🙂 And so I call it Scientology in his honour 🙂
That’s a puzzler. GPS is a product of science. And so are computers and computer programs.
What is it that you claim cannot be properly explained by science?
I agree with Dan, that science does not explain everything of interest. But I’m not seeing how your GPS example illustrates that.
Neil,
“What is it that you claim cannot be properly explained by science?”
You seem not to have read my comments as a whole.
Neil,
I have already spelled out my claims in quite a lot of detail.
I did. And you are saying that science cannot explain your reasons for your behavior. So why mention GPS?
Neil,
“ So why mention GPS?”
I think that is really, really obvious from my comments. I just don’t see the point in making the same arguments again.
Dear Labnut. Let’s take the famous example of 11 moving dots being presented to observers, who are asked about the age, sex, motor skills, and mental state of the human whose movements have been abstracted from a movie as the trajectories of key joints and extremities (point-light animation). Naive observers are easily able to diagnose happiness, depression and so forth from short clips. They have a window into the subjective world of another based on a prior, obviously quite high-level, “database” of observations that they have correlated with behaviours of themselves and others.
For your problem of interpreting GPS tracks we can do exactly the same thing – compare tracks of other runners, travel ourselves over your route and record it, observe your actual footprints etc etc. Then we have enough information to model your algorithms nicely. Stepping back, what are the deep motivations for you being a runner – how much do we learn from studying large groups of runners as opposed to your introspective reports? How much information do we gain from looking at discretionary exercise in other animals eg do mice like running in a treadmill in the same way you like running? I don’t think it is reductionistic to say that humans do many things other animals do, but to excess, and to speculate this is an intrinsic feature of having a big brain (not a feature of being minded or having culture). When looking on youtube recently for washing machine repair instructions, I found numerous videos (looking into the tub) of an entire wash-rinse cycle of particular older machines. A pretty good “reductionistic” psychological explanation is that a percentage of the population are at the extreme of a particular personality dimension and happened to work in a particular industry, combined with the concept that rhythmic repetitive sounds and movements can be relaxing. A “manifest image” explanation probably wouldn’t be that different, but the scientific image, psychological style, is to assess the importance of these different factors, maybe combining this with genetic data on personality type, maybe looking at extremes of this personality type in the psychiatric diagnoses of personality disorder (which are at least partly “just” deviations from social norms that cause distress in self or others). This is the thing about psychology qua science, everybody thinks they understand all this already – it’s all just common sense i’nnit.
Re intensional versus extensional language, it occurs to me that on the one side the foundation of the intentional is a transcendental form of thought (e.g. Heidegger). The foundation of the extensional narrative has become, of necessity, science. At the present time the extensional narrative is in the ascendance. One can I expect this trend to continue as long as science finds new, wonderful and important things that we were previously ignorant of.
In practice, however, the two forms of communication and description are inextricably linked.
Hi David:
The comparison between Collingwood and Darwin is a good one. They are both looking for the best explanation. Here’s how I see the difference.
Collingwood is trying the reconstruct the reasoning of Roman military engineers is a particular set of circumstances. He asks himself, if the wall is really a sentry-walk, and not like a town-wall, then what else would we expect to find? He answers in terms of what would seem to him to be reasonable if he were in that situation. He can then check whether the archaeology bears out his reasoning about the Romans’ reasoning. It is a rational inquiry into a rational enterprise. (He may also have documentary evidence of how Roman military engineers thought, but this is not essential.)
Finch’s beaks not being the product of intelligent design, the problem for Darwin is to account for them being fit for purpose. They exhibit diversity and this correlates with location. This can only be so if the beaks are changeable and adaptive without intelligent input. So his problem is to explain how something can look like a functional artefact without actually being so. This requires him to put together what he knows about variation (from pigeon-breeding) and competition. Reasoning about reasoning doesn’t come into it.
Alan
I think animal symbolicum is making a good comment above. “Only if we represent an event as an action does it make sense to ask about the reasons for the action. I was pointing out that we don’t have to know the reasons why someone acts in order to represent him as acting.” The point I would add is that it is intentions that make actions actions.
Of course not all intentional acts turn out as intended. But even then intentions remain part of the action.
Take two cases. Case one, A intends to shoot B and does so successfully. Case two, C intends to shoot D but fails and shoots E instead. From one viewpoint the two action types are identical: A shot B, C shot E, they are both shootings.
But is there no difference between them? Of course there is: A did what he intended, while C did something she didn’t intend. “A shot B” is a correct description; “C shot E” is also correct; but neither of these descriptions is complete. The complete description requires adding the intention. In the second case it is “C shot E while intending to shoot D”. Adding the intention adds a counterfactual: “C would have shot D and would not have shot E were it not for some misapprehension or mishap”. And in the first case “A shot B” is incomplete until we add that the shooting was intentional.
Intention is a necessary category in the human “sciences” whenever we are talking about actions and not just events.
Hi Alan. I kind of think that the teleological properties of reason derive (at least partly) from the teleonomy of being alive. This is very much the space the enactivists are in when they say stuff like “living is cognition, even for bacteria”. At a higher level, even just in the sense that if reason is the slave of the emotions, these derive their purposiveness from evolution. So biologists _can_ usefully think about what they would do if they were solving a particular problem of living, and consider how evolution has got “stuck” because it can’t jump over to a better solution on the next peak. In interactions between different species say, we meaningfully talk about “arms races”, “cunning subversion” eg cuckoos, toxoplasmosis.
One more little thing to keep in mind: there is the intentional, the extensional and the unintentional (unconscious mind).
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2440575/
David: I have no problem with that. We have the sort of reason that goes with being big-brained language-using weak-bodied pair-bonding social mammals. Alan