The multiple realizability thesis (MRT) is the idea that the function of certain kinds of objects can manifest itself in a number of different ways. In Multiple Realizations, Shapiro points out that in order to properly understand the MRT, one must first distinguish between the two different kinds of objects: natural kinds and non-natural kinds. This is relevant to MRT because, as Shapiro points out, only natural kinds are multiply realizable. Moreover, natural kinds are “defined by their purpose or capacity or contribution to some end” (643). A natural kind is a term used in natural science. This means that a natural kind is a kind that is defined by its essence. An example of this is a corkscrew. A corkscrew is characterized by its essence to open bottles. Furthermore, a non-natural kind is a kind that is not defined in virtue of its essence. A non-natural kind is a term that is not objectively agreed upon; it is an object that does not fit into an objective category. Moreover, a non-natural kind is a term that is not used in science.
In the beginning of his essay, Multiple Realizations, Shapiro clarifies what is and is not multiply realizable via an example of corkscrews. Shapiro first shows that two corkscrews that differ only in their color are not multiple realizations of a corkscrew because the two different colored corkscrews do not “differ in their causally relevant properties” (644). This means that in order for two different objects to be multiple realizable, the two objects need to contain their own, unique properties that contribute to the essence, or function, of said object. In the case of the two different colored corkscrews, the only differing property of the two corkscrews is the color. There are no properties that are unique to either corkscrew that contributes to the corkscrew qua corkscrew. Thus, the two different colored corkscrews are not multiple realizations of the kind corkscrew. However, Shapiro points out that two completely different corkscrews, such as a waiter’s corkscrew and a winged corkscrew are multiple realizations of the kind corkscrew. This is because a waiter’s corkscrew and a winged corkscrew “differ in their causally relevant properties (644). Since the function of the waiter’s corkscrew and the winged corkscrew is brought about in two different ways, (i.e. CO2 injection and manually screwing), because the properties that contribute to the function of the two corkscrews are completely different, the two types of corkscrews, waiter’s corkscrew and winged corkscrew, are multiple realizations of a corkscrew.
Furthermore, Shapiro seems to contradict himself on page 647 of Multiple Realizations when he questions the “coherence” of MRT when he says:
The problem is … Either the realizing kinds truly differ in their causally relevant properties, or the do not. If they do not, then we do not have a legitimate case of multiple realizably, and MRT, in the given instance, is false. If the realizing kinds do genuinely differ in their causally relevant properties, then, it seems, they are different kinds. But if they are not different kinds, then they are not the same kind, and so we do not have a case in which a single kind has multiple realizations.
In this passage, Shapiro is pointing out that, if his interpretation of MRT is correct, then it implies that there can be no similarity between two or more objects of the same kind other than the function of said objects. This is where the distinction between natural kinds and non-natural kinds comes into play. Again, a natural kind is a term used in natural science. Moreover, a natural kind is a predicate that can be used in a scientific law. In order to be a predicate, the thing being talked about must be categorical – it must fall into some sort of taxonomy. For example: eyes. What makes human eyes, animal’s eyes, robotic eyes, and other eyes, similar to one another, is the fact that each thing called ‘eyes’ fits into an objective, agreed upon category called ‘eyes’. This taxonomy of ‘eyes’ can then be used in scientific inquiry in an attempt to learn to ‘science of vision’ (653). In theory, this is important because, though they are many differences between human’s eyes, animal’s eyes, robotic eyes, and others, there are, presumably, some similarities. Namely the capacity to see. Moreover, philosophers, such as Jerry Fodor, believe these similarities might be able to uncover some laws that pertain to the ‘science of vision’. Shapiro disagrees with this notion, though, because these laws would concern special science, and Shapiro does not think there can be any laws of special science. According to Shapiro, whatever laws that could be derived from the apparent similarities between humans eyes, animal’s eyes, and robotic eyes, are “empty” (649) because there is nothing in common between these three eyes other than their capacity to see.
According to Shapiro, whatever laws that could be derived from the apparent similarities between human’s eyes, animal’s eyes, and robotic eyes, is “empty” (649) because there is nothing in common between these three eyes other than their capacity to see. If there are no similarities between the three eyes besides the capacity to see, then the only law one can derive from the similarity between the three is ‘an eye has the capacity to see’. If this is the only law that can be derived, then this sole law is not very informative.
Since the laws that can be derived from categorizing certain kinds as natural kinds are “analytical, rather than empirical” (646), they lose their pragmatic importance. This is because they no longer contain any empirical information. As a result, the relevance/importance of the special sciences, as put forth by Shapiro, disappears, as well. This is due to the fact that Shapiro’s idea behind the special sciences was to “provide laws about functional kinds” (653). However, since Shapiro has shown that there are no laws that unite functional kinds, (natural kinds), the goal of the special sciences — which is to provide unifying, empirical laws for natural kinds — must be reevaluated. As a result, Shapiro suggests “we should see the special sciences as pursuing an understanding of how functional kinds produce the capacities that make them interesting” (654). This means that the special sciences should be used to help understand the function of natural kinds. Shapiro thinks that the special sciences can help shed light upon why it is that a natural kind’s function is what it is.
By distinguishing the difference(s) between natural kinds and non-natural kinds, Shapiro shows that MRT is possible for natural kinds only. A kind is a natural kind if and only if it is an object that is defined by its own essence. An example of a natural kind is an eye. An eye is defined by its essence – its ability to see. Moreover, eyes are “particular realizations of a functional kind” (654). This means that eyes, which are defined by their ability to see, are realized by any capacity whose function is ‘to see’. Moreover, these capacities can only be different realizations of a kind (to see) if and only if the kind – specifically, the properties that contribute to the ability to see – are different. Therefore, there is nothing in common between two or more realizations of eyes other than their function ‘to see’. As a result of this breakthrough by Shapiro, the role of special science has changed and has become nothing more than a way to better understand the different types of natural kinds.
You can read Shapiro’s full article here.
Shapiro, Lawrence A. “Multiple Realizations.” The Journal of Philosophy, vol. 97, no. 12, 2000, pp. 635–654. http://www.jstor.org/stable/2678460.