Functionalism in the Philosophy of Science
It's not often that a philosopher comes along with the claim that a concept he's been developing fits right in at the foundation of a discipline -- that it's not just a new development exploring some fringe of inquiry, but shores up the entire field.
That's what I think one of the key functionalist concepts of my metaphysical functionalism does.
Here's the funny thing. This concept itself isn't revolutionary; I think it's simple and obvious. But, after years of searching the philosophical literature, I couldn't find anyone who had hit upon it. Many philosophers have used it and talked all around it, but no one I've found ever said "this is a fundamental concept for the philosophy of science". And that's what I think has revolutionary potential here: the realization that this concept can help us structure our beliefs about the world in an astoundingly productive way.
Now, for quite a while I worried that my idea was too obvious. Could it be that philosophers all understood it, and never mentioned it because it was so elementary?
No. Philosophers, if they do anything at all, dig down to the most fundamental levels of analysis and explanation to make explicit the core concepts that can make sense of our world. Anyone who had found this would have made a very big deal of it, I decided. After all, that's what I'm doing....
And now, without further ado, the concept I've been talking about is ... valence. 'Valence' is a dispositional term -- which I probably should explain for the lay reader. A disposition is a variety of property (i.e., a trait or an attribute or a quality belonging to something) -- specifically, it's a tendency to behave in a particular way under certain conditions. Probably the favorite philosopher's illustration is this: sugar has a disposition to dissolve in water (at room temperature, etc.).
Now, maybe you're so quick that you noticed right away that there's something different about dispositions as compared to other kinds of properties something might have. Sugar has a number of static properties, such as being white, or granular, or sweet -- things that characterize it under normal conditions. But this disposition of solubility ... well, it's not static at all! It's a functional property -- one that describes how sugar works under special circumstances.
Now, I'm sure you'll understand why a functionalist might be more interested in dispositions than in static properties. But consider this: might there be a variety of disposition that is even more dynamic -- more functional -- than properties (like solubility) of substances?
Yes, indeedy! Valence at your service!
Valence is (by my definition) a disposition of a situation rather than of an object. What's a situation? Well, it's not much more than your ordinary understanding of the term: a dynamic array of objects and forces existing in a fairly well defined (normally local) environment, causally isolated (for all practical purposes) from external influences. In other words, a situation includes all the objects and influences you need to know about (and no "excess baggage") if you want to predict what might happen in that situation.
Strictly speaking, though, a "situation" is a fiction. There will always be external influences operating on the things in a local environment, and it's usually impossible to tell whether the non-local influences will be significant. But there's really no difference between talking as though a situation is independent of external influences (having defined the term so that internal influences determine the outcome of a situation) and talking as though an object is separate from its environment. Objects, too, are fictions -- for science tells us that you can never really establish a definite boundary between the alleged 'object' and its surroundings. [Is this molecule part of the object, or has it evaporated into the surrounding atmosphere? Is this electron part of an atom belonging to the object, or has it migrated to a neighboring atom? These are problems for a static view of objects -- but if you're a functionalist, you can treat objects as dynamic systems that tend to interact with their surroundings as a unit. Same with a situation: it, by definition, is functionally isolated from external influences to the extent that its behavior is internally determined.]
Now that you understand better what a situation is, let's delve into the nature of valences. Maybe you remember from high school chemistry that this term 'valence' is used for a tendency of an atom to add (or, alternatively, share) a particular number of electrons belonging to other atoms. The valence of an atom tells you what sorts of bonds it is likely to make (given the opportunity) with other atoms. To say that sodium (Na) has a valence of +1 says volumes more than to say that sugar is white. It implies, all at once, that we can expect to find in the world things like salt (NaCl), sodium fluoride (NaF), and a whole host of other compounds!
I (blush) stole the term 'valence' from the chemists. Sorry, but it wasn't doing as much as it could while restricted to chemistry. So I liberated it.
If I have my way, 'valence' will forevermore have a larger meaning. It won't mean just a disposition of atoms to bond to additional electrons. It will mean any tendency of the objects and forces composing what I call a 'situation'.
Here are a few examples of valences for a variety of situations, just to give you the general flavor of how ubiquitous valences are in the world:
|an electron approaching a naked proton
||a hydrogen atom forming
|a buyer with money and a seller with a product or service
||a deal (or no deal)
|a bacterial infection and your immune system
||forming antibodies to the bacteria
|a candidate you like running for office, and you in a voting booth
||a vote for her
|fleas on your dog
||a whole lot of scratching going on
|a hungry cheetah near an antelope calf on the veldt
|a helium balloon released into the air at ground level
||up, up, and away!
|a website you find interesting, and you're online
||more clicking and reading
Now, a structuralist might object that these are cases of electrons tending to orbit protons, or of buyers to buy products, or of candidates to attract votes, etc.
Um ... or is that protons tending to capture electrons, or sellers tending to vend, or voters tending to pick their favorite candidates, etc.?
See? Neither side of the transaction in these examples is wholly responsible for the outcome. They function together as a situation -- and the property in question (the disposition to behave in a particular way) really belongs to the situation rather than to any of its constituents. Some valences give the illusion of one-sided action; for example, it may seem that the helium balloon is just passively pushed upward by the surrounding air -- when in fact it is shoving aside the air above it as it rises.
I think you get the idea -- and maybe this will clench it: When Sir Isaac Newton formulated his Second Law of Motion, he said that "For every action there is an equal and opposite reaction." This means that you start with an action, and then there's a reaction, right? Wrong. I think Newton understood quite well that the "reaction" was simultaneous with the "action." But
the world view at the time was a structuralist one, obsessed with agents acting on the objects in their environment. Indeed, under the Aristotelian view still prevailing at the time, even objects were treated as agents that "wanted" (for example) to reach the lowest point possible -- the pre-Newtonian conception of gravity. Newton's legacy, then, invited a conception of physical interactions as being composed of sequential events: action followed by reaction -- a sadly mistaken notion, given that in reality action and reaction are two faces of the same coin: interaction -- which results from the very general sort of arrangement I call 'a situation', rather than from one of its components unilaterally "acting on" another.
OK, now you know what a valence is. But what good is it for the philosophy of science?
Well, I think there are several important contributions it can make. One is that it provides a common conceptual foundation for all the sciences. Why? Because each science is really about examining certain varieties of valences. Thought of this way, they no longer seem to be so isolated from one another. For the school of philosophers of science who believe that the role of science is to order and systematize our knowledge -- to make inquiry into the phenomena of our world more economical and fruitful -- this should be a welcome contribution.
Furthermore, it becomes easier for scientists working in related disciplines (say, virology and cytology) to recognize that they don't need to fight over which of their fields "owns" the study of the "borderline" phenomena between their disciplines. To understand how a virus invades a cell, it's not enough to consider the action of the virus on its environment, or the action of the cell; one must study the valence between virus and cell.
Another important contribution is that the concept of valence enforces a functionalist perspective. It steers scientists away from the humanly natural, yet scientifically mistaken view that they are studying the actions of agents, the behaviors of objects, etc. This structuralist perspective, I think, obscures the functional nature of the situations in which we find what looks like agents and objects. Such situations are better understood as arrays of components of interaction. Structuralism robs those who study causal situations of a clear understanding of their problem.
Probably most important to the philosophy of science, though, is what valence can do for explanation. The dominant school of philosophers of science today regards the job of science as explaining the phenomena we observe. A leading theory (offered by Peter Railton) is that what constitutes scientific explanation is a complete causal story of the influences leading up to the event observed. This improbably complex story, called 'the ideal explanatory text', traces the causes of causes of causes ... all the way back to the Big Bang. Obviously, we're not likely to be able to construct the ideal explanatory text for much of anything -- it's far too complex. But even if we treat this as an unreachable ideal -- something at which to aim without any expectation of complete success -- that still leaves us wondering how we can get a real-world, everyday handle on scientific explanation. Some sort of approach is needed that will allow scientists to explain events in such a way that all the relevant causes are included, yet the explanation isn't impossibly unwieldy -- and valence, I think, can do the job.
Why? Well, events emerge from situations -- and if we use my definition that a situation is a dynamic system whose behavior is determined by the interplay of internal objects and forces, then we have limited what we need to explain enough (in most cases, I think) to actually produce a satisfying explanation that includes all relevant factors and excludes irrelevant ones. Yes, we will be ignoring external factors that actually do have an influence on the behavior of the components of the situation in question, but only those external factors whose influence is not great enough to change any such behavior in any way that makes a practical difference. Indeed, if an allegedly external factor exerted enough influence to change the behavior of the system, then it must by definition be included in the situation!
I'd better make it clear that valence in itself isn't an explanation! That would be like saying, "This situation tends to produce a particular sort of result because of ... Ta-da! ... a tendency!" That sort of claim would justifiably have philosophers rolling in the aisles.
Instead, what valence gives us is the general form of causal explanations of any variety. It does so in ways that I'll bet will make sense to you: It includes all the pertinent causal influences (both those we know about and those we don't), excludes extraneous factors, and requires a causal "story" that can take you from those influences to the observed (or predicted) result(s).
Explanation, then, becomes a matter of
- specifying the components of a situation, and then
- describing their causal interplay in such a way that the actual outcome makes sense.
Pretty slick, eh?
There's another important contribution that valence can make to the philosophy of science, and it lies right at the core of the discipline. What is the GOAL of science? That question continues to be debated. Is the goal, as one view I've already mentioned claims, to provide explanations for the events we see in the world? Or is the real goal to unify our concepts of what there is in the universe and how it works? These are the leading candidates competing for the attention of philosophers of science.
Not being particularly shy, I'll propose another aim for science -- one that I think better represents the psychology of the human organism. Are you ready?
The goal of science is to "zero in" on the optimum valence between our potential for understanding and the facts of the universe.
In other words, the universe is the way it is -- and for each identifiable feature or function of stuff in the universe, somewhere in the world of ideas is the best match between that feature and some conception about it. We should realize that maybe some limitation of human psychology could prevent us from actually entertaining the optimal conception in some cases, but even so, there would still be some conception that is available to our psychology that "fits" the pertinent aspect of the real world better than any alternatives -- and that is the valence we can hope to fill.
Karl Popper presented a general model that fits this perspective quite nicely: his Three Worlds model. Think of the external, physical world of material things as World One. Next, consider the world of your internal mental experience -- which is, after all (most philosophers since Hume have agreed) the only connection we have to World One. This world of inner experience is World Two. Finally, we can talk about the world of abstract entities: numbers, sets, propositions, linguistic terms, general conceptions, musical themes -- anything that might take up residence in your (or my) World Two, but is in principle available to anyone whose train of thought takes him there. This world of abstract entities -- a very Platonic notion -- is World Three.
Now, I should mention that some philosophers disapprove of claims that abstract entities really, actually, for real exist. After all, have you ever seen the number 42 floating by? I'm not talking about the numerals '42' or 'XLII', or the union of six groups of seven physical objects. These are physical, World One representations or instantiations of the abstract entity 42, which itself has no material existence in the physical world. Maybe you can see, then, why some philosophers turn up their noses at the notion that we can ascribe actual existence to such things. Next thing you know, they might complain, we'll be crediting such things as unicorns and sakes with bona fide existence.
Popper's division of worlds helps a lot here. We don't have to claim that abstract entities have existence in World One, along with tables and chairs. They're World Three critters -- and maybe you or I can lure them into our personal World Two for fun or profit.
Still, some philosophers will prefer to avoid this kind of talk. Hilary Putnam, for example, has suggested that we can avoid spooky claims of existence for mathematical "objects" if we talk instead about possibilities. We don't need to say that 2 exists in some Platonic realm in order to do math. We can simply consider all the possible ways things might be logically arranged in pairs! Each such arrangement (or all of them taken together) might be what we humans really mean by '2'.
Personally, I think that claims of existence for possibilities is even spookier than those for numbers. I also think humans can't avoid using abstract entities, so there's not much point in denying their existence. But we needn't argue over the point, for I don't much care whether people are prepared to believe in the actual existence of World Three entities like the generic concept of tree, as long as they accept that you and I can both have World Two mental states whose content is appropriately described as there's a tree, and those mental states might normally be generated by our interaction with something in World One that we'd agree to call 'a tree'.
OK, back to the main point. Let's think of these three worlds as forming a sort of ecology. When you think tree and I think tree, we'll find it beneficial if we have both zeroed in on a general World Three concept, tree, which allows us to be thinking and talking about the same idea. Perhaps even more useful would be the existence of a TREE in World One that is causally responsible for producing, through our senses, the tree thought for each of us.
Let's translate this into my own terminology. There is a valence for us to use the same term (in our personal World Two) to correspond to the same World Three concept. This allows us to communicate. Furthermore, there is a valence for each individual to develop a World Two concept (or apply one already developed -- perhaps received from the World Three subset that is our culture) to each World One object he encounters. This allows us to think.
Extrapolation from these beginnings gives us science, which requires not only that we reliably form our own World Two representation of a World One object, but that we express it in World Three terms that can induce the appropriate World Two representation, in someone else, of the self-same World One object.
Now, if I were reading that for the first time, I'd need to go back and reread it a few more times in order to make sure I understood it. So, don't feel bad if you feel the need to digest this a bit at this stage!
I'd like to wrap up this section with a few comments on how metaphysical functionalism can help with the ongoing realist/anti-realist debate in the philosophy of science.
This will be too simple an account to satisfy all (or even most) philosophers, but it should suffice to give you an idea of the key disagreements here. This isn't a fight about whether or not the material world is real; existence of a material world is generally accepted. The questions are whether, when we express propositions "about" the material world, we are actually referring to something outside our own minds; whether such propositions can be true or false; or whether we can have knowledge about the material world. Realists tend to affirm these notions, and anti-realists tend to deny them.
I'm sure there are other ways to frame this debate, but these will do for now. I just want you to see what Popper's Three Worlds model, together with my notion of valence, can do to smooth things out for the disputants.
First, let's look at reference. When we think there's a tree, does that thought refer to an actual, material thing? If not, have we failed to make an adequate connection between our thought and the material world? Well, we have a World Three concept, there's a tree, which has happily taken up residence in our own personal World Two. To be precise, the World Three concept is instantiated in our World Two, in the form of a thought with content there's a tree. Other minds have access to that concept, so it's not just a matter of personal ideosyncracy. But does that concept refer to a real-world tree?
I say it does. Reference isn't some magical connection between thought and material reality, and furthermore it doesn't require a perfect grasp (in thought) of reality. Bumbling is allowed. Reference can be vague, confused, or even mistaken, and yet refer.
I think reference has two elements:
- A thought refers to a material entity if it is (at least partially, and usually primarily) caused by interaction with that entity; and
- A thought refers to a material entity if it is intentional with respect to ("about") that entity.
I think philosophers have placed too much emphasis on the latter of these factors, focusing on the human component rather than the environmental one.
We now pause in mid-thought for lack of time to continue....
Please Be Patient....
This page is still under construction, so I
apologize for the incomplete synopsis of my views. This page is, however, probably the fastest growing part of the website ... so if you check it periodically you're likely to encounter new material once in a while.
Why not continue on to Part 5 of the Overview to see some other implications of Metaphysical Functionalism for the various branches of philosophy?
Thanks for visiting the Evolution and Philosophy website, and I hope you find it provocative and stimulating. And don't forget to
leave your comments for me (and others) on the Message Board!