The Multiple Realization Definition

Foreword

This paper is my last paper for Philos 951: Larry's Favorite Papers. The title is a misnomer, I am not defining multiple realizability theory. Rather, I argue that they way in which some words are defined practically necessitate that they be multiple realizable. It seemed like such a clear concept to me, however it seems very rarely touched on. I argue that because of these definitions, we may prefer a non-MRT based metaphysical approach to objects. However, I have not taken a metaphysics class and I may be mistaken. Nonetheless, I hope this provides an interesting viewpoint. One should go in with good knowledge on MRT, what MRT responds to, and objections it may face.

Body

Multiple realizability theory (MRT) is a pretty clear concept: there are many ways for the type mousetrap to be tokened. It is hard to refute this in any productive way. However, reductionism has always had a strong grasp on my intuitions. In this paper, I wish to argue that the way we have defined some words presumes that MRT is true, and our intuitions of why MRT is true are based on those examples. The goal here is not to necessitate that MRT is wrong, but rather to argue that there are different types of definitions, and which ones we use to help us solve this metaphysical puzzle requires investigation.

Consider common examples of things that are considered to be multiply realizable: mousetraps, corkscrews, hearts, etc. It is immediately apparent that these things are multiple realizable. They are essentially defined by what they do. A mousetrap kills mice, a corkscrew unscrews the cork from a wine bottle, and the heart pumps blood around the body. Though it is of interest exactly what the purpose of each of these things is, we know there is some function that they are meant to serve to be that thing. However, if these things are defined functionally, it seems to necessitate MRT. There are an endless number of ways for a mousetrap to be realized, thus, it is pretty clear that MRT must be true for these sorts of things, without arbitrarily hedging on the idea that there must be some underlying physical mechanism that necessitates these objects. I wish to promote the argument that the definition of these words is not necessarily sufficient to disprove reductionism.

However, consider the concept of temperature. It is currently defined as ‘the degree or intensity of heat present in a substance or object’, and scientifically it ‘reflects the average kinetic energy of the vibrating and colliding atoms making up a substance’. Despite the temperature of an object being epistemically understood as the reflection of average kinetic energy, it was initially understood subjectively. Some possible definition, like ‘As an indicator of the "hotness" or "coldness" of an object relative to a standard’, seems like a good description. This keyly differs from the idea of temperature now, there are times in which something feels warmer or cooler without it actually being the case, such as putting your hand out of the window of a driving car. Someone at this point in time may have wanted to claim that the temperature of a gust of wind was really colder than what they were initially feeling. Assuming this initial definition and the scientific knowledge of the time, it would be very hard to hedge a bet that there is some scientific underlying way for all expressions of temperature to have some fundamental physical explanation. Especially because it is not the case that the gust of wind is cooler than the temperature.

Further in history, a thermometer was invented to have a more standard way of measuring temperature. This updated tool clarifies that what we feel as temperature differs from the real temperature. However, even with the thermometer, we still end up unable to get the temperature of solids. It again would remain unclear at the time that there may be some necessary physical realization of what the temperature really is. It would seem that the attribute temperature simply had multiple ways to be realized.

Consider now the discovery that the temperature of a gas is directly proportional to the average mean molecular kinetic energy of the gas. Now, we have an epistemic and specific physical backing for the definition of temperature for a gas. The kinetic theory of matter was initially postulated specifically for gases: it was easier to understand at this level. At the time of the discovery of this rule, it was not yet known whether the same idea of temperature could also apply to solids or liquids. Assume now that it turned out that it wasn’t the case that the average mean molecular kinetic energy of a solid or liquid had a direct correlation to its temperature. I know this intuition may be hard to throw out, but at the time of the discovery, this would not have been a bizarre thing to assume.

To me, it seems like the temperature of a gas and the temperature of a solid or liquid should now strictly speaking, become distinct concepts. Imagine it one day turned out that, instead of kinetic energy, the temperature of solids and liquids was proportional to the average spin rate of the molecules that make up the matter. Imagine a textbook that did not differentiate these two concepts. They have fundamentally different properties and ways of calculation, the ideas need to be differentiated.

However, consider the non-scientific definition of temperature, where we consider it just a feeling of reference to hotness or coldness. Using this concept alone, one can seemingly easily call these two distinct concepts realized tokens of the type temperature.

I argue that many other types of words necessitate something similar. If we define some word in similar functional fashions, it is clear that we can muster multiple ways to realize these words. This becomes even easier to do when we have these large and abstract concepts, such as pain. Our current knowledge of pain seems analogous to our previous understandings of temperature. If we are justified in saying that the discovery of those strict types of temperature should, strictly speaking, be different concepts, something similar can be said about pain.

As I have touched on in every paper, this makes no claim about the usefulness of these generalizations. Just as in our hypothetical, it would still be useful to think of those concepts as being similar enough for ordinary language; these overarching ideas of pain would still be useful for psychology.

One may simply want to claim that these types of temperatures are indeed realizations of temperature. This is a perfectly reasonable intuition, and the point of the paper. Though I may wish to disagree with that idea, I can’t say much against it. I would like to point out that the progression of how we understood temperature, from being a subjective feeling, then progressing to an epistemic question about what the reading would be under a thermometer, seems to imply that we prefer definitions that have more epistemic justifications. My guess is that in the hypothetical world, we would end up distinguishing these two concepts. However, I still cannot say much against it. The argument being provided is that understanding temperature as being defined in that functional way allows it to be realized in those functional ways, as opposed to being understood as a fact-of-the-matter issue. These differences in definition guide whether or not we think they can be multiple realized. Which one is the better way of defining is up to further argumentation.

Overall, it seems like some words guide our intuition towards wanting such tokens of the type to be multiply realizable. However, assuming knowledge of strict rules that make up its components, that intuition seems to dissolve. There are valid arguments for the MRT proponents to get out of this, this is not damning against the position whatsoever. However, what type of definition that should make up our metaphysical understandings of the world should be investigated.