First those of you who don’t already know of this may enjoy Philosophy Bites, the short podcast. While I find more logical fallacy in the arguments than I thought I would, it is still an interesting weekly look into the work of today’s philosophers.
The most recent one is on Human Agency i/e free will.
Now I am compelled to believe that there is free will. Of all the empirical evidence I have the evidence that I act freely (subjective maybe) seems to me to make any claim that I am not a free agent the extraordinary claim that requires the extraordinary proof. Further I’ll say that we do not know nearly enough about ether physics or the brain/mind to be convincing we are right in the causality problems. Just one (albeit contentious) of many possible solutions that allow free will is the ORC-OR hypothesis. There are many others and it seems to me that nether side is close to solving the puzzle, so any position has to come down to belief.
And I think Jennifer Hornsby, in this interview was agreeing with me (so it gives me pleasure as confirmation bias often does) but she gets into some very twisty language in the middle, and after three listens I am not sure what she actually thinks to be true.
Doug, can you follow her reasoning (though I expect you will disagree with it)?
OK, I’ve gone and listened to it ... the audio isn’t very long. It will come as no surprise that I’m a proponent of what Hornsby called the “standard” view in philosophy of mind, whereby actions are caused by belief and desire states. To be fair to Hornsby, there isn’t enough time in this short a piece for her to explain precisely what her alternative view is, nor for her to argue what she sees as the problems with the standard view, nor for her to argue why her view does not fall afoul of those same problems. I vaguely recall reading some of her stuff many years ago in a grad class on action theory, but my recollections are not precise enough for me to reconstruct anything more. Let me just say that I find her position baffling, since either it is just another way of formulating the “standard” view, or it is obscuring what is really going on at the physical level.
She says that actions are caused by persons. We all agree with that. She also says that her view of actions has them caused by reasons, so that we are acting intentionally—i.e. with some intended end in view. But the standard view agrees with all of that. When she says that the woman crosses the road to catch the bus, the standard view agrees; her reason for crossing the road comes through the content of her desire. Why did she cross the road? Because she desired to catch the bus. Sometimes we linguistically shorten this by saying, “Because the bus was on the other side.” There isn’t any real difference there.
Now, I believe that Hornsby is mostly a philosopher of language, so I expect that her view is going to be influenced quite a lot by the particular linguistic descriptions we give of things, in particular of action events. IMHO this sort of approach to philosophy is very wrong-headed. We cannot get at the reality of things in the world simply by looking at the linguistic form of a description. This is the worst sort of armchair philosophy. Fortunately this sort of approach, a hold-over from the early part of the 20th c., is slowly going out of style now. Instead of looking to our parochial linguistic practices, we should instead be looking at the sciences. Hornsby only mentions “physics” when talking about the sciences, which is a subtle method of straw-manning her opponents. Yes, physics is the queen of the sciences, and in some sense all causal phenomena must be cashed out in physical terms. But that said, the appropriate sciences when it comes to actions will be at a much higher level: neurophysiology and psychology, for example.
By implying that somehow her picture is not consistent with a physicalistic picture of the causal antecedents to human behavior, she runs the risk of seeming frankly antiscientific. Does she really believe that there will be no worked-out causal picture of human behavior in terms of neurophysiology? Or that neurophysiology at base biochemistry? Or that biochemistry is at base physics? I expect not, but then it simply isn’t clear how her picture is any different from the standard view of action, except perhaps in a manner of speaking. (That is, she’s describing it using different terminology without actually falsifying any of its main tenets).
We cannot get at the reality of things in the world simply by looking at the linguistic form of a description.
I think this might be similar to what she seems to say about science: scientific statements are descriptions about the world, and they cannot get at the reality of things. It will always be an approximation, much abbreviated and simplified, and will not be able to take everything into account, even in a simple case of crossing a street to catch a bus.
She seems to be saying that people are the locus of some kind of extra cause (free will?) in a causal chain, i.e., that there is something going on in people that cannot be explained (though perhaps it can be described) by the causal model used in physics.
I agree 100% with Hornsby. I admire her work and her courage for standing up against a standard philosophical orthodoxy. Sometimes orthodoxies can be wrong; and their very entrenchment unfortunately leads to the continued propagation of bad ideas—for example the Cartesian “ghost in the machine” view, of which I find the picture of agency that Hornsby criticizes akin.
I have tried to articulate a similar sort of approach (in opposition to the “standard story”) elsewhere in these threads here and have basically found universal opposition. I finally gave up arguing for the alternative story. I think that some people tend to think of pictures that differ from the standard story of agency as anti-naturalist or anti-scientific (as evidenced above). I, of course, beg to differ—that’s why I call myself a “pragmatic naturalist.”
Rather than deciding in advance that there is nothing but the sort of causation that physics tells us about, which leads to a monistic reductionism (of all things in terms of mechanical, inanimate cause-effect relations) and a feeling of mysteriousness when it comes to concepts like “freedom,” “agency,” and “life,” I leave other options on the table. The most plausible option, I think, is of the sort that Hornsby offers. It is basically pluralistic about natural relations. There are two sorts of relations in nature: non-agential, inanimate, merely causal happenings and the agential doings and sufferings of living beings. Inanimate stuff doesn’t do anything. In fact I don’t think that evolution would be possible if it was all purely mechanical all the way down—if nothing did anything but stuff just happened. Evolution requires beings that do things to their environment in an effort to obtain energy from it and survive to reproduce. Life is special, not because it is animated by an immaterial soul (this is the misconception that many, I think, apply to deviations from the standard story), but because it is alive. Distinguishing living things from non-living things is not anti-scientific or anti-naturalist.
There are two sorts of relations in nature: non-agential, inanimate, merely causal happenings and the agential doings and sufferings of living beings. Inanimate stuff doesn’t do anything. In fact I don’t think that evolution would be possible if it was all purely mechanical all the way down—if nothing did anything but stuff just happened. Evolution requires beings that do things to their environment in an effort to obtain energy from it and survive to reproduce. Life is special, not because it is animated by an immaterial soul (this is the misconception that many, I think, apply to deviations from the standard story), but because it is alive. Distinguishing living things from non-living things is not anti-scientific or anti-naturalist.
Well, for one thing, agency is not a thing—like a “life force” or a “soul”—that could potentially be passed back-and-forth or transferred from one thing to another, now animating this person, now that, now animating this animal, now that rock. There is nothing spooky or non-natural about this use of “agency”—it requires only that we not place the causal relation as the only relation by which one can explain natural occurrences. Agency, in this sense, is merely a characterization or classification of a property of certain natural things in which the persistence of those things requires that they acquire and utilize energy to maintain themselves and propagate—they do things to their environment and suffer the consequences of those actions. See Erwin Schrodinger’s “What is Life?” for a fuller explanation of this particular natural characteristic of things (you can Google title and author along with “pdf”).
Here is a pithy quote from a Wiki entry on organism to describe the essential point I am making (and, I take it, that Hornsby is making too):
Organisms are complex chemical systems, organized in ways that promote reproduction and some measure of sustainability or survival. The molecular phenomena of chemistry are fundamental in understanding organisms, but it is a philosophical error (reductionism) to reduce organismal biology to mere chemistry. It is generally the phenomena of entire organisms that determine their fitness to an environment and therefore the survivability of their DNA based genes.
And, here’s one from the Wiki article on Vitalism:
In terms of the biology of the cell, a variation of vitalism can be recognized in contemporary molecular biology; for example in the proposal that some key organising and structuring features of organisms, perhaps including even life itself, are examples of emergent processes in which complexity arises out of the interactions of the chemical processes which occur in the cell; When individual chemical processes form interconnected feedback cycles which produce products perpetuating these cycles rather than unconnected products, they can form systems with properties that the reactions, taken individually, lack.
Whether emergent system properties should be characterized with traditional vitalist concepts is a matter of semantic controversy.
it requires only that we not place the causal relation as the only relation by which one can explain natural occurrences
This is a really interesting, and I think really challenging question. Certain properties of complex systems only make sense at the level of the system as a whole, and living organisms are a great example of this. This is, I think, what the concept of emergent properties is trying to acknowledge. I’m not sure you can prove this means extreme reductionism is wrong, only that it is more useful for certain kinds of analysis to view complex systems as a whole rather than an assemlage of parts. Even if the whole really is just the sum of the parts (and I’m not sure when this is true or when it isn’t or how exactly we would know), the whole is often what we’re interested in and certain questions make no sense unless we take a more holistic view.
Still, the quote above seems to stretch beyond this to the idea that the properties of matter, laws of physics, and interactions of material components in a system are not sufficient to generate the properties of the system, and that “agency” is an additional something needed to explain these properties. I don’t think this is truly what you mean, based on the rest of your post, but I can see how the line between this and supernaturalist vitalism gets pretty fuzzy when you employ language like this. Are you saying complex systems (or just living organisms?) cannot be only the result of the interactions among their components, generating perhaps properties none of the components have but doing so by the nature of their own properties and interactions, or are you only saying that some things about such systms cannot be understood only by analysing the components and their interactions?
Are you saying complex systems (or just living organisms?) cannot be only the result of the interactions among their components, generating perhaps properties none of the components have but doing so by the nature of their own properties and interactions, or are you only saying that some things about such systms cannot be understood only by analysing the components and their interactions?
I’m not sure if I fully understand the question you are asking here, but hopefully the following at least responds to it.
I don’t doubt at all that the living things we see around us today evolved from much simpler forms of life and ultimately from single-celled organisms, which themselves evolved from simpler molecules. Of course I’m not alone in this. We send probes to Mars to test the soil to see if it contains water because the presence of water is one of the essential environmental conditions for the processes of life to take root. At what point do we call these processes “Life”? My suggestion, or better, speculation, is that we have “life” when such simple cells or molecules begin to do things to their environment—manipulating it or moving around in it—in order to acquire energy sources to continue the processes that maintain it. In my view these are the roots not only of life but of agency as well—so it’s not as if we’ve got two separate things here: living things and agency; rather they arrive on the scene together. In ordinary language we tend to use the term “agent” to refer to intentional agents—and some think that the term “agent” applies only to conscious, language-using human beings alone. My use of the term “agent” (and Hornsby’s too, I think) is much broader. The roots of our agency go way back.
Agency is something that one can observe, it is not some immaterial, spooky life-force invisibly animating things. I think that this is the point that Hornsby was getting at when she said we can describe the behavior of the woman crossing the street simply by saying she wanted to take the 59-street bus. She did that; she was not caused to do it—as if, either by some mysterious, invisible force or by some infinitely long chain of determined causal connections. Thus agential relations are just as observable as ordinary causal relations and so perfectly natural.
I don’t really see myself as making any grand metaphysical claims here. Essentially, I’m suggesting that we need to bring two different approaches to our study of nature. There’s nothing wrong with physics and there is nothing wrong with describing things in terms of mechanical cause-effect relations. It’s just that that’s not all there is. There are naturalists like Newton and Naturalists like Darwin (non-overlapping magisteria, if you will). Schrödinger, for example, was puzzled by the fact that living systems tend to maintain (potentially infinitely) a high degree of order—even, it seems, an increase in order. This fact seemed to conflict with the (second) law of increasing entropy. (As an aside, Schrödinger’s “What is Life?” is what inspired Watson and Crick to the discovery of DNA.) Noticing this conflict is what led Schrödinger to introduce the concept of “negative entropy” to describe how living things can be distinguished from non-living things.
I’ve re-read your question (more than twice) and think I understand it better. So hopefully I can answer your question more straightforwardly:
If I understand your question correctly, then I reject the first part (of the disjunction) and accept the second.
Both complex systems and living organisms (which are one species of complex systems) are, in my view, the “result” “of the interactions among their components”. Of course a lot hangs on the meaning of “result”. The phrase that follows that one (viz. “generating perhaps properties none of the components have but doing so by the nature of their own properties and interactions”) leads me to suspect that by “result” you mean something along the lines of emergence. I think I’m okay with saying agency is an “emergent property”; however I don’t think that applying that term provides much explanatory value.
I am saying that these “systems,” which we call “agents,” and their interactions and relations, “cannot be fully understood only by analysing the components” of the system. One must also look at the whole.
I would not use Schrödinger as a base here; his popularized and speculative work on biology is not mainstream. Going back to my post above, I specifically rejected the approach that we must cash out all our causal relations in terms of physics. Such work will almost certainly always be beyond our limited abilities. Instead we will cash out behaviors in terms of properties of the nervous system, including most importantly the brain. This, on one level of description. On a higher level of description we can cash out these same behaviors in psychological terms, as the standard view does explicitly in terms of beliefs and desires—which are not themselves properties found in physics.
Your rejection of vitalism doesn’t really capture what vitalism is—the claim that the difference between living and non-living things is captured in terms of a property (so called “élan vital”) which is not itself reducible to properties at any lower level. This amounts to the claim that there can be no science of biochemistry, since biological organisms would not simply be complex assemblages of chemicals, but rather assemblages of chemicals animated by some mysterious “élan vital”. I hope and expect you will agree with me that this sort of vitalist picture is pseudoscience through-and-through and must be rejected in the firmest of terms. Biological entities just are complex assemblages of chemicals, and knowing chemical processes such as the Krebs cycle allows us to know how those chemicals are “animated”; able to move about and do things.
As an aside, the quote you provided on “organisms” from wiki is irrelevant to the point I have just made about biochemistry. “Fitness” is a property of biological organisms that has to do with their relation to the environment; although once again this could, in principle, be reducible to chemical properties of the environment, in practice such reduction will always prove out of our reach. We simply lack the capacities to do this. But this has nothing to do with the main point which is that all biological processes are themselves constituted by chemical processes. (The property of “fitness” is not a biological process, but rather a relation between the biological organism and its environment).
This is basically a discussion of emergence. When and why does a bundle of acids become an organism, at what point does the I in this bag of cells emerge, how does a group of unconnected bags of cells make things like markets, or seemingly without anything other than chaotic independent “decisions” make any city street a ballet of foot traffic? Asteroids in an asteroid belt following simple physics crash into each other and bounce around in a chaotic mess…same with gas molecules.
I find that it does not require anti-naturalism, but an admission that naturalistic physics has not yet explained all that much. And physics (just as one example) only explains things in terms of perfect models. Even our physical models get beaten to crap every time we actually start to thrown in the dynamism of real world situations. Take the Space Probe Anomaly (not to be anomaly hunting) as one example of this, but there are many.
However being a bunch of monkeys we readily defend our pet theory even when it is not all that well understood or supported by evidence. To say that since we have observed nothing but causal chains of gas molecules moving in a vacuum jar, therefore all the universe works on the same principal, is quite a stretch, and flies in the face of the every day very real appearance of choice and independent action - agency, that is evidenced a billion times a day (albeit not scientifically). Just because the simple mathematical model fits the former doesn’t mean we haven’t missed a very big equation in there some where.
And just because the complexity of studying a real world phenomenon makes it currently nearly impossible to study does not mean that the simplistic perfect-state study and finding should rule as fact, it is at best a best guess hypothesis in as much need of further study and proof as the other (I would argue more, given the group experience as strong evidence of a kind).
Thanks for your response. I have to claim a position of agnositicism on the question of whether or not complex systems can theoretically be understood in meaningful ways by looking at causal chains traced back to the level of chemistry or physics. Doug seems to think this is possible in theory, though perhaps not in practice, meaning that such causal chains are the genesis of all the phenomena we observe at the system level even though we may never be capable of tracing all the necessary connections. You and Chris seem to be of the opinion that some properties of such a system can only be understood at the system level, by applying the principle of agency, and that no degree of sophistication of analysis would ever be sufficient to understand the poperties of complex systems solely by reductionist analysis of their components because some properties only exist at the system level. I tend to lean towards the latter view, but I doubt it could be proven definatively either way. Our simplified models have proven very powerful for understanding and manipulating complex systems, so the reductionist paradigm must have some validity because it works. On the other hand, I work with living things and I am personally certain that they are not practically comprehensible in a purely reductionist way, and I am doubtful that they are even theoretically comprehensible in such a way. Even if they are, it is often more efficient to view them at a higher level. I don’t. of course, see a necessary antagonism between these two views. Reductionism is a useful tool, holism is a practical device for daily use, and the level at which the question is asked will generally tell us the level at which we should look for the answer. I prefer a holistic approach by temperment, but I can’t deny the enormous value of reductionist approaches to my work, so I think there’s room for both so long as neither gets carried away and begins to demand total primacy.
Certain properties of complex systems only make sense at the level of the system as a whole, and living organisms are a great example of this.
Do you think that these properties are real in the sense that they exist independently of your conceiving them?
Agency is something that one can observe, it is not some immaterial, spooky life-force invisibly animating things.
Same question, only substitute agency for properties.
In this paragraph:
“Fitness” is a property of biological organisms that has to do with their relation to the environment; although once again this could, in principle, be reducible to chemical properties of the environment, in practice such reduction will always prove out of our reach. We simply lack the capacities to do this. But this has nothing to do with the main point which is that all biological processes are themselves constituted by chemical processes. (The property of “fitness” is not a biological process, but rather a relation between the biological organism and its environment).
there seems to be much that depends on how these properties are real.
There are limits to the accuracy of our knowledge about the properties of the most fundamental parts of these systems, the properties of particles, and these cannot be transcended. Any useful knowledge of the actual properties of actual systems will tend to be statistical, which I think approaches the kind of abstraction of these ideas of agency (or even of something as vague as elan vital).
So, we have to take into account ineluctable limits on the knowledge we can have of a state that can comprise or ground a cause or effect, and that the more we know about such a state, the more we change it. We can only observe the statistical behaviour of systems, not only in practice, because our tools are dull, but in theory, because of uncertainty; and, worse, the more detailed your observations, the more you must alter what you observe.
So, there is no real possibility, theoretical or practical, of knowledge at that fundamental level in the way that reductionism posits.