On reductionism

All the talk on reductionism as of late has kept my mind quite occupied. So occupied, in fact, that I felt the need to devote an entire post to reductionism itself. Though reductionism does come in different forms, this post will deal with the form that claims that an object is nothing but the structure of its physical constituents. This form of reductionism is very appealing to naturalists for the following reason. If the natural world is all that exists, then objects are nothing but physical particles in different structural forms. Hence, a lump of bronze and a bronze statue consist of the same matter simply arranged differently.

The initial difficulty with this view was articulated perfectly by attentive commenter JD Walters:

[D]escribing a cat as a mere 'amalgam of physical properties' is the (metaphysical) bundle theory of particulars and has all sorts of problems, including the consequence that if even one of the cat's properties changed, even the number of hairs on its body, we would be dealing with a different cat.

Let us expound JD’s argument here. Let’s take a cat, which on reductionism is nothing but a hunk of matter arranged cat-wise. This cat, then, is defined by the structure and arrangement that the particles are exhibiting. So, now let us remove a single hair from its body. What do we now have? Well we now have a cat with a different physical arrangement than before, and thus we have a different cat altogether. Why? Because cat A (the first cat), though it looks the same as cat B (the second cat), has more physical constituents than cat B—the difference between their constituents being the particles that make up a hair—and thus has a different physical arrangement than B. Hence, since cat A’s matter has a different structure and arrangement than cat B’s matter, then cat A and cat B are not the same cat, on reductionism.

Now, this certainly seems to run counter to our common sense—though the fact that it runs counter to common sense only demonstrates that reductionism, then, runs counter to common sense. For when a cat goes for a haircut, say, we don’t really believe that the post-haircut cat is a different cat than the pre-haircut cat. But, this only means that there must be something more to a cat than its mere physical constituents.

However, one attempted rebuttal that I’ve heard so far from reductionists is that the reason we still consider cat A the same as cat B is because our brains still categorize that there are defining characteristics or properties in cat B that make it the same as cat A. But this response runs into problems. First, how our brains characterize objects is superfluous to the discussion at hand. Talk about how our brains characterize objects is epistemological, as well as psychological, while talk of what actually makes an object an object in mind-independent reality is ontological. Hence the above objection is only conflating ontology with epistemology. To reiterate, when we talk of reductionism we’re talking about what makes an object that very object—i.e., what makes a cat a cat, as opposed to a frog. Thus stated, we’re interested in substances in themselves, and not how we perceive or categorize those substances.

Second, this objection, even if we ignore its metaphysical conflation, still does absolutely nothing to answer the original point. Remember that on reductionism an object is nothing but the arrangement or structure of its physical constituents. So if cat A has a different structure (even by a few particles) than Cat B, then, on reductionism, they cannot possibly be the same object. And thus it makes no difference whatsoever how we categorize said objects, because they are necessarily different objects because they have different physical constituents and structures!

Furthermore, the topic of how we categorize objects, based on an ontology of reductionism, leads to further absurdities in itself. Let us return to cat A and cat B. It is true that the difference (a hair) between A and B is miniscule when we perceive it, and therefore our visual perceptions will regard A and B as the same. But what if the difference is greater? What if cat B is also missing a leg? Can cat B be recognized as cat A? Well, not in the way the reductionist would want, since, again, cat B has different (less) physical constituents than cat A, and such constituents have a radically different arrangement than cat A as well. But suppose, contra the conclusions of reductionism, that the reductionist still wants to affirm B and A to be the same cat. Well, we can go even further. Suppose cat B is missing a tail, an eye, and an ear, in addition to its missing leg. Suppose the cat continues to deteriorate. At exactly what point is this cat B, on reductionism, no longer cat A? Notice here that the reductionist would be stuck in a dilemma, because at whatever point he identifies that the physical arrangement of cat B no longer constitutes cat A, the only argument for stopping at that specific point, consistent with reductionism, can be that cat B’s arrangement or structure is different from cat A’s—because, again,  on reductionism an object is defined as nothing but its physical constituents. Now if the reductionist wants to claim instead that cat B is different than cat A because it no longer retains the properties that make it cat-like, then the reductionist has given up reductionism. For admitting that there are properties inherent in a substance or object that make it that very substance is to give up reductionism. That is, it is to admit that there is something in a substance over and above its physical structure that makes it a certain substance.  

But we can push even further against reductionism, by returning to our original difference (one hair) between cat A and cat B. Notice that if cat B has the same structure as cat A minus one hair then cat B’s structure and arrangement is subsumed under the structure and arrangement of cat A. That is to say, cat A’s arrangement contains the arrangement of cat B. Now we can continue this process by removing two, three, and four hairs etc., to create a cat C, D, and E etc. Notice that each subsequent cat is contained in the former ones. That is to say, cat A contains all of cat B, C, D, and E etc. But this leads to something quite peculiar. Since, on reductionism, all of these cats have different structures and arrangements then they are all different cats. But this means that cat A literally contains cat B, C, D and E etc., which means that present in cat A is a multitude of other different cats at the same time in the same place! Such is the absurdity of reductionism.

Reductionism, then, turns out to not only go against common sense but, also, to lead to absurdity. Obviously, then, an object or substance cannot be nothing but its physical constituents. Thus stated, there must be something more to substances in themselves over and above their parts.

 

Essentialism and reductionism

Mike D over at The Aunicornist has, once again, devoted a couple of posts to refuting the metaphysics of Thomism. Since I am myself a Thomist I felt the need to deal with such arguments and assertions. I will, for now, be dealing with this post of Mikes. So, let’s dig in.

Mike states the following:

[W]e all agree that if tomorrow all humans were wiped off the face of the Earth,cats (for example) would still be cats—that is, they would retain the amalgam of physical properties that our brains categorizeas simply “cats”, even there [sic] wouldn’t be anyone around to say, “Hey, that’s a cat!” But does it follow, then, that cats have a property of identity that makes them cats?

Here Mike demonstrates his poor understanding of Thomism by conflating identity with essence. What Mike is talking about here is essence which is just that whereby something is what it is. To grasp a thing’s essence is to grasp all the properties that are essential to that thing remaining that very thing. Or, to take Mike’s example, the essence of a cat is comprised of whatever properties a cat must have for it to remain a cat. Yet, Mike here uses the terms “the property of identity.” But, identity is not the same thing as essence. It is an essence that makes something what it is and not identity. Identity is more broad and more tautological. The property of identity simply means that something is identical to itself and is not identical to what it isn’t. A corollary of this is that all of a thing’s properties are contained in its identity. Yet this is not the same with regards to a thing’s essence.

Let’s use an example to illuminate the distinction between identity and essence. Take two chairs. Each particular chair is identical to itself and is not identical to the other chair even if they both look  exactly the same. Yet, these chairs do share the same essence in that they both provide a seat—here providing a seat is the essence of being a chair. So, two objects can share an essence without being identical. This is because identity takes all of a thing’s properties into account whereas essence only takes essential properties into account. It is essential that a chair provide a seat, but it is not essential for a chair to have four legs. Yet, when contemplating a chair’s identity it is crucial to include the property of how many legs it has—for a chair with four legs is obviously not identical to a chair with three.

All this is important because Mike, in the quote above, is again conflating identity with essence, which wouldn’t have happened if Mike adequately understood the position he’s attempting to refute. So, while none of this, so far, undermines Mike’s arguments, it does demonstrate that Mike does not understand his opponent’s positions the way he thinks he does.

Mike then illustrates his (already confused) argument with an example of a mountain:

Mountains are formed, if I remember grade school geology correctly, when massive tectonic plates press against each other, forcing the earth to slowly rise over eons. I wonder how the Thomist might think about this, then—at what point does the earth have the property of identity of a mountain, versus just being a really big hill or a giant pile of rocks?

Again, Mike should be talking about essence here, and not identity. For a mountain, or hill, at any stage in its formation will still always be identical to itself. And thus, the “property of identity” will always be present. What Mike means to argue here is that if we watched the earth slowly rising over eons then at what point could we say that the earth exhibits the essence of a mountain, as opposed to a hill? The point of this example is to demonstrate that determining the essence of something is not always so clear cut, and that such an endeavor can run into ambiguities and vagueness.

Yet, this argument doesn’t really make the case that Mike thinks it does. First, a Thomist would not claim that determining a thing’s essence is always an easy endeavor. It is in some cases very difficult to determine what the essence of something is, or how such an essence differs from the essence of another. But nothing about this difficulty demonstrates that there are no such things as essences. The real problem then deals with epistemology, and not ontology. Encountering difficulty in determining a thing’s essence is an epistemic problem, while proclaiming that things do not have an essence is an ontological problem, and Mike is conflating the two.  In order for Mike’s example to have any force he would need to demonstrate that difficulty in determining a thing’s essence entails that essences are therefore nonexistent. But Mike has not done this.

Second, Mike’s argument actually makes the case for Thomistic essentialism, and not against. By picking a vague case of determining essence, Mike is implicitly conceding that there are cases that are not vague. In fact, most cases are not vague. One could easily identify the essence of the brain, the heart, water, human rationality etc. The fact is that we can find an abundance of easily identifiable essences all around us. So why should we call essentialism into question because of a few exceptional cases that run into obscurity, instead of affirming essentialism due to the abundance of its applicability? There is no reason. The fact is that we wouldn’t even be able to recognize vague instances of identifying essence if we didn’t already have clear and precise examples.

Mike then articulates one of his central arguments:

Thomists take things like identity, essence, nature, 'prime matter' and potentiality to be literally real properties of the external world, independently of human minds. But at every turn, we can see that we have no reason whatsoever to think that any of these 'metaphysical' properties are anything more than conceptual constructs. There's no reason for us to think that the concept of "cat" is anything more than a useful categorization of our brains for a particular arrangement of matter; we have no reason to think that there exists any such a thing as the identity, essence, or nature of a cat independently of our minds.

Mike is making the argument that our metaphysical concepts are mind-dependent, and that they are simply constructs of our brain that help us order and identify the world we experience. There is no essence of a cat independently of our minds. The essence of a cat is simply something that our brains have constructed in order to arrange the matter that makes up a cat.
While this viewpoint might seem parsimonious, it is rife with problems. First, the world is exactly the way we would expect if there were in fact actual essences and identities of things. One example of this is unity. Let’s illustrate. If I consider all four of my dogs—my Chihuahua, Dachshund, Beagle and Blue Heeler—I can see that all of them share the same essence of “dogness” even though they are not identical. And in that sense they share in a unity between them. They are all individual particulars yet they are in fact related to one another. They are related to one another in a way that they are not related to a cup, a tree, or a lamp. But if Mike is correct, and dogs share no “essence” and we simply construct such things in our brain, then why is it the case that we attribute a shared essence to all these dogs? The only way one can answer this is to say that these dogs, in mind-independent reality, exhibit similar characteristics and have similar properties that allow us to group them into these related “kinds”. But then this simply makes the case for the Thomist! For this is what the Thomist has been arguing all along. For dogs can only be grouped into a specific kind because they have certain dispositions and properties that make them dogs, as opposed to, say, frogs. And it is these dispositions and properties that constitute an essence. So, the fact that the world contains a multitude of particulars yet many of these particulars are unified only makes sense on essentialism.

Second, Mike seems to forget that our concepts are always derived from reality itself. He is so smitten with Model-Dependent Realism and Embodied Realism that he seems to think that our brains literally make this stuff  (like essences) up. (A common example he uses is the way our brain “creates” color. But this is false. Our brains don’t create color. They simply perceive the raw materials of sense datum in a certain way so as to “see’ color. But everything that causes us to see color (e.g. wavelengths and photons ) already exists in the world. So color does exist, it simply exists virtually instead of formally. All this is to say that we perceive color because color exists virtually in reality, and is not a simple construct of our brain. ) But if our concepts are abstracted from reality then this simply isn’t the case. To use an example, I can tell the difference between an apple and an orange. Why? Is it simply because my brain arranges the matter into two different categories? Or is it because there is a real difference between the properties and dispositions of an apple and an orange and my mind is abstracting those differences? Obviously it must be the latter. But if this is true, then there must be something about an apple that makes it different than an orange. And we know this to be true because my understanding of the essence of an orange is different from the essence of an apple. But these essences weren’t simply constructed by my brain, rather they are rooted in the properties of the objects themselves in mind-independent reality.

Mike then states the following:

[W]e can reject the Thomistic metaphysical gobbledygook on the principle of parsimony — the notion that cats have a distinct, non-physical property of 'catness' (their 'essence'), for example, is completely superfluous to our understanding and description of what a cat is. We can have a fully accurate, useful description of the animal simply by recognizing it as an amalgam of physical properties which our brains categorize in a particular way, and nothing more. There is no need to postulate any extra non-physical or 'metaphysical' properties to understand what a cat is, why it behaves as it does, or what it evolved from. Since the assumption of the existence of such things is not essential to our description or understanding of cats, we can discard it. We don't even have to demonstrate its falsity — i.e., somehow 'disprove' the existence of those metaphysical properties — we can simply discard them as superfluous and thus meaningless.

A few responses here. First, nobody has argued that postulating things like essences and identities are pragmatic. So why would Mike be looking for pragmatism here? Why? Because Mike has an empiricist presupposition, and anything that is not scientifically observable, predictable, and measurable is, to him, not pragmatic and therefore “superfluous.” But in order for Mike’s position to be vindicated, he would need to demonstrate that science is the only path to knowledge. But Mike knows that this is self-undermining. So why, then, should we believe that metaphysical concepts like essence and identity can be thrown out simply because science has no use of them? The answer is that there is no reason.

 Moreover, let it not be forgotten that science is only interested in the quantitative aspects of the world and not the qualitative. Yet, metaphysical concepts like identity and essence are not quantitative, and therefore the fact that they might not be pragmatic in scientific investigation is completely irrelevant, since science isn’t looking for them to begin with!

Second, Mike’s argument above implicitly affirms reductionism. That is, Mike believes that any entity or object is nothing more than an amalgam of its physical parts and components. But reductionism runs into problems. For example, if water is nothing but hydrogen and oxygen composed together, then water should not have any dispositions and properties different from its constituents. But surely water does have properties that differ from hydrogen and oxygen. So then water cannot be nothing more than its physical parts. At this point one might say that we should expect water to have different properties than its parts alone, because water isn’t simply hydrogen and oxygen, but hydrogen and oxygen bonded together. So of course hydrogen and oxygen bonded together will behave different than hydrogen and oxygen by themselves. But this is exactly the anti-reductionist’s point. If water behaves differently than its constituents, then there is something about water as a substance in itself that amounts to more than its parts. Thus stated, if a substance has properties that its physical parts do not have by themselves, then there is something over and above these parts alone that make up the dispositions of said substance, and therefore a substance cannot be said to be nothing but an amalgam of its parts.

Hence we see that Mike D has failed to make his case. First, he has shown that he doesn’t even correctly comprehend the position he is arguing against. He constantly rebukes those who recommend Scholastic literature to him, yet he’s shown that his research into this very area is minimal and is lacking. So why not at least buy a book that gives a formal defense of such a position? That would be my recommendation. Second, he has tried to demonstrate why metaphysical concepts like essence and identity are meaningless and superfluous, yet his arguments are rife with problems. Third, he’s attempted to affirm metaphysical theories like reductionism which, when considered, only lead back to the very metaphysical concepts he was attempting to overthrow. I maintain, then, that nowhere have we seen good reason to overthrow Scholastic metaphysics, let alone metaphysics themselves.