Natural kinds by tommaso tonello
Now, this has made some essentialists claim that essentialism perhaps adapt better to certain types of science/part of a science rather than others. Other essentialists went in the direction of rethinking essentialism in a different way, in a way that is able to handle at least some of these objections. There are 2 main amendments of essentialism in the literature in philosophy of science (they are not only 2, but they are the main). The first one is called “property cluster theory” of natural kinds: this is a more relaxed view about natural kinds, for it only requires members of a kind to share a subset of properties that might end up clustering together due to some underlying cause. There are different types of properties: some properties are more proper to assemble than others. These are actually properties that cluster together, and it’s enough for a kind to target these properties in order to say “well, such and such is an instance of that kind”. There is no necessity, in other words, that the properties are alway the same, the properties that identify a kind are always the same, but it is enough that some of them are shared in the way that that kind describes. Take again the case of biological species: members of a species may share many (not all) properties caused by various mechanisms to belong to the species (for example, sharing a common ancestor is one candidate, or sharing a particular ecological mesh, or coming under a particular mechanism of development or a particular gene exchange. Some of these properties can get to cluster together, but then this leaves open a wider possibility of variation of difference. So, the clustering is a more relaxed criterion than essence in the way we’ve been talking about. The other version of essentialism goes under the name of “promiscuous realism”. The philosophy of science/biology in question (that advanced this theory) is J. Duprè. In Duprè’s perspective, essentialism is part of what we could call and epistemological path that also includes 2 Reuther perspectives that are important to essentialism in the same way in which essentialism is important to them: determinism and reductionism. Now, all together, these 3 perspectives (determinism, reductionism and essentialism) build up what we could describe as a united image of science, or science under a unified image. But, according to Duprè, this is an image we must resist. In his book “The “disunity” os science” (1983) he argues in favor of a disunity image/in favor of the disunity of science. In this book, Duprè argues against essentialism (for he believes that we always deal with a plurality of classification of reality), against reductionism (there is not only one level of reality that is more basic than any other, which all others ought to be reduced to, it’s not just underlying physics and then everything is reduced to that), but instead we have to think of reality in terms of different levels of descriptions each with its own reality/purpose, and then he also argues against determinism, or as he calls/cause it, we are also epistemological monism, namely there is only one single methodology that gives us a unique criterion of what it mens to be scientific. What we find in science instead is a plurality of methods, of epistemic virtues that are used in the picking of our theories, and what we take to be our best theories in certain circumstances. That’s the kind of scalpel of the argument. Although Duprè is totally against determinism and reductionism, he is not totally against essentialism, and that’s why his version of what he calls promiscuous realism is actually kept alive but it’s been amended in such a way that he can actually sort of sit (?) In a revised version of what he thinks science is about. Let’s see what this promiscuous realism is all about. The premise from which Duprè formulates his position is metaphysical: science, according to him, doesn’t progress if it doesn’t operate on the basis of strong assumptions concerning the world it investigate (the causes, the entities, the movements of objects and so on). The metaphysics of modern science is taken to be deterministic instead, in the sense it is governed or it is set to be governed by universal laws which order reality in such a way that there is alway a materialistic foundation to anything and to which everything can always be reduced. There is an image that represents this at its best, and that’s the metaphor/image of the clock: the image of a mechanism that once set in motion works in a kind of perfect and precise manner, in a stable and determined way. Now, it is true that more contemporary versions of the cosmic clock have become very complex/ sophisticated, sometimes it has been told that it even goes beyond human comprehension. For example, think of Chaos theory: we are told here that just because of this kind of intrinsic complexity of how things work it has become also impossible/very difficult to make precise predictions. Even in these extreme cases, says Duprè, basic metaphysics really is not questioned. So, this idea of order depicted in the way we just described (by reducing everything to a materialistic basis, the determinism that underlines laws of nature and so on) stays the same. The metaphor of the clock is useful in particular to understand the thesis of reductionism. To understand how a clock works, it is necessary to understand how its constitutive parts actually work and function and then equally important is to understand how they interact/they work together. We start from the larger components of the watch, the main wheel, the case and so on, and then we go down to the more basic bits and pieces of the mechanism. And even further, we go to the mechanical properties of each of the bits and pieces. At that point, we kind of create a sort of hierarchy that goes from the complex to the simple, and the tendency is to reduce the complex to the more and more simple components. The same works for science, and this is precisely the type of image that has dominated the idea of modern science, this is how scientific theories were set to work. Again, reductionism and determinism go hand in hand, it is easy to understand why. Take the behavior of atoms, these minute components of matter: they must be supposed to be deterministic if they must work as the foundation of all the rest. Something cannot change the way it is if it has to provide the bedrock for how all the rest actually works. Now, central to this mechanism/mechanistic perspective is essentialism as well, and interesting for us, Duprè discusses essentialism precisely in the context of natural kinds, which he believes is the version that is most debated in contemporary philosophy of science as in the details. So, here is the way Duprè presents the view on natural kinds and his own view on natural kinds: again, this is the starting point for the background. “Besides individual things, nature is constituted of types of things which organize individual things (?) (that’s the picture we’ve been trying to put forward us)”. For example, what the world comes to us not only with individual pieces of irons or individual human beings, but with so-called “iron things” or “human things”. To be iron or to be human has an objective claim to reality. At the same level, we’re prepared to say that there are things that are made of iron. So, what have natural kinds to do with the mechanistic image/the clock? The fact that something is a clock more or less determines what types its constitutive parts must belong to. To be a clock hand is different from being, say, a human hand, or to be a clock hand is different from being a clock wheel. SO, within the same mechanism, parts and components organize themselves in a kind of hierarchy and then, thanks to the fact that different parts belong to different kinds, then the mechanism actually works. We can indeed talk in terms of a mechanism: if we cannot distinguish a hand from the wheel, then there is no mechanism to talk about. Now, this idea that things belong in a non-ambiguous way to kinds is clearly connected with essentialism, for the reason we’ve been discussing, and essentialist perspective tells us that what turns something into a kind of a thing is precisely the fact that this thing possesses its own property essentially, that is properties that are necessary and sufficient (“if and only if”) to making these things belong to its kind. So, these properties are perfectly added clues (?) to legitimate the question “what kind does this thing belong to?”. A perfect position to us to ask that question. As we mentioned earlier, we said that, unlikely reductionism and determinism, Duprè is prepared to defend some kind of essentialism. So, he believes that we are entitled to support the view that there are some basic divisions that organize things in the world. He is prepared to defend it. But, he says, this basic divisions are not enough. We need far more classification than the basic one. Also, it is not always that clearcut what kinds/kinds of things are things belong to. And again, he is a philosopher of biology, so he knows what he is talking about. And often, the belonging can only be decided, according to him, by context of use, for context of use specify the reason/reasons why we need to classify things in a particular way, or why we need to classify something in the way we do. This sometimes makes things fuzzy, not that clearcut, but this doesn’t mean, according to Duprè, that there are no natural kinds, but it becomes important if not mandatory to ask ourselves what contexts make things belong to one kind rather than another. So, in a sense, we could say Duprè is against strong essentialism, and he argues that there are different ways equally valid in their own context of dividing up the world, and that’s why this theory becomes his perspective, becomes promiscuous. There is a promiscuity of kinds in his vision. But he just concentrate on the promiscuous side.
How does this promiscuous position position itself vis-à-vis the more traditional maths and physics that we’ve been describing? According to a more traditional metaphysics, classifications/natural kinds must first of all reveal one orderly arrangements of the things that exist, and this means that it must be possible in principle to decide on what side of a classification a thing locates itself (either in or out, in a clearcut manner). Secondly, classification must be discovered, not invented. This means that a kind, in order to be natural, must belong to nature, must belong to the furniture of the world before becoming a kind for science. Science is in the position of discovering the kinds that are in nature. And then, thirdly, correct classifications/natural kinds tell us everything that we need to know about an object, everything that is essential. So, a natural kind must include everything that is fundamental for the kind to be that kind. This is spelled out in terms of intrinsic/essential properties of properties that make something what it is in an essential sense either than fundamental sense. Now, as we said, Duprè, in order to sort of save some kind of an essentialist view of natural kinds, needs to sort of get out of this type of metaphysics, that is too narrow and brings us in direction that we cannot handle, so he needs to think of rescuing essentialism by putting it in a different metaphysical perspective. How does he do that? How does it proceed? The starting point is what he calls the “ontology of common sense”, namely what we do in ordinary life/circumstances. And the ontology of common sense is indeed intrinsically pluralistic. Common sense imposes meanings onto a world which is otherwise confused/chaotic and so on by using classifications which are sometimes simple, sometimes even inaccurate, not particularly “to the point”, and then little by little, by increasing our knowledge, we are able to refine this categories and make them more precise/accurate. Our scientific knowledge often puts us in the position of modifying some of our common sense classification, give them the accuracy that they lack. For example, from the everyday classification of fish, now we know that we cannot include whales. So, still, this inaccurate steams allow us to, sort to speak, baptize a kind, they give us access to a class of things, and then we are able to refine the access depending on our growing knowledge. Here is the catch: this doesn’t mean that the original classification must be abandoned. Common sense classifications and scientific classifications are just different. Common sense classifications are identified/put forward with particular goals/uses in mind: for example, they reflect particular anthropocentric visions, whereas scientific classifications are made/put together/assembled with other goals, goals that perhaps do not/should not include any anthropocentric element, they should be more objective. Take the case of a picture of beautiful flowers (lilies). Now, the lily family is more inclusive than what we are used to refer to when using the term “lily”. So, ordinarily, when we talk about lilies, we only refer to the flowers, and we exclude onions and garlic that still belong to the same family. But this doesn’t make us conclude that the classification of lilies as only as flowers is vague and unreal, and only a classification that includes flowers, garlic and other types bulbs is actually more objective and reliable: they simply serve different purposes. In ordinary speech, we might just focus on the flower, whereas we exclude everything that has to do with culinary purpose. But this does not make our classification of lilies less natural. It still picks something pretty essential about the kind. In this way, we are not denying essentialism about kinds, we are denying that essentialism has to go necessarily in the direction of unity/completeness. So, different constructions of natural kinds should be allowed because this idea of different constructions doesn’t deny that all these different kinds actually exist, it only denies that one construction is able to support all the uses we put these kinds to. We cannot reduce all the different types/kinds to one. Duprè is a philosopher of biology. He knows, for example, that there are at least 3 ways to classify species, and accordingly to establish whether an individual belongs to one species rather than another: there is a morphological way, and evolutionary way, a physiological way, but each of these criteria of classification is valid, and in fact it can demonstrate its validity in context. If we are after the evolutionary history of species, this is what justifies using a classification of that sort. So, it depends on what we’re looking for, in what we intend to identify about species. This is not a move against essentialism or realism of kinds. Back to our fish example: some entity can be a real whale, a real mammal, a real predator and even a real fish in some context. So, many if not all of these classifications somehow describe/designate appropriately a certain object depending to the context where we intend this classification to do its work. To be an essentialist/
realist (we’ll tell more about the word “realism” in a moment) does not mean shrinking our ontology necessarily, shrinking the ontology in which we feel entitled to believe, but instead acknowledging the potential richness of an ontology, especially when it come to the ontology of science. Of course, we might think this is a form of essentialism that is far too liberal, for ultimately any categorization that is at least minimally grounded on some aspect or other of the world ultimately can qualify being a natural kind, and this seems in plain contradiction with the underlying central intuition of essentialism, namely the fact that essential properties allow to identify specific groupings in the world which are grounded in and by nature, under, (also, the intuition related to this) that by being discovered by science, these entities should be superior to any common sense classification that we make when we speak within our community/daily life. But the promiscuous view, according to Duprè, does not deny that scientific classification might be better for certain uses: for example, it might be more accurate in terms of its explanatory power. He only denies that common sense classifications should be declared inferior by comparison. We shouldn’t compare them, they are just suitable to different uses, they have different goals in mind. Of course, choosing a common sense classification requires a specific justification of why we stick to it. And this puts us in the position of having to resort and go back to our interests/goals and so on, we have to come up with a reason why we’re using it, but this is per se not a reason against accepting them in our picture. Why is all this discussion about natural kinds so important to scientific enquiry, and why so many debates are centered on accepting or rejecting the existence of natural kinds? We need to expand further our horizon of discussion and include a word that we just mentioned so far: realism. In particular, scientific realism. Scientific realism is the view that there is a way that the world is, independently of us, and also that there is a way (sometimes we can say different way) in which we can have access to this world. One way to access this world is by our scientific theories. And in principle, and it might happens as well in practice, we have reason to believe that at least some of our best theories might manage to successfully represent this world, or at least the part of the world the they focus on. Natural kinds are what our theories are about. So, a metaphysical treatment of natural kinds secure, to some extent, reliable reference to our theories. And we say this both in the ontological and the epistemological sense. In the ontological sense for we can claim our theories are about the real world, what is out there, independent of us on our mind-making efforts, and it is epistemological for natural kinds are ultimately classifications, so they are produced by our theories, and the reason why we produce them is because our theories provide an epistemic access to the world. That’s why realism and natural kinds, at least according to some views, go hand in hand. It is a corollary of scientific realism that when all goes well the classifications and taxonomies employed by science correspond to the real kinds in nature. Well, how this correspondence can be actually matched out is a complicated matter.
We asked a question, whether there are justifications classifications provided by science/our scientific theory that allow us to sort of say that our theories are actually about nature/they capture something about nature that is really there (carving nature at its joints). How do we take the right properties? One of the possibility is using the notions of “essence”. The so-called essentialists about natural kinds say that a kind is natural if it corresponds to a mind- independent grouping that exists in nature. Therefore, natural kinds and realism often go hand in hand. For example: it is a corollary of scientific realism that when all goes well the classifications and taxonomies employed by science correspond to the real kind in nature (again, this idea of correspondence). Correspondence is not an easy relation to be picked, especially not in this context. How can we account that a kind as classified by our theories (they are kinds like electrons or quarks) corresponds to some real entity in nature? How can we envisage this kind of correspondence? For a start, theories are linguistic descriptions, whereas the world is made up of events/facts/ entities/material properties. What is the correspondence really about? What do we put in correspondence? Besides, often the properties that we refer to in our kinds and properties that purportedly correspond to real properties are not really observable, they are precisely the result of some theoretical speculation (we don’t observe electrons and their properties). For that reason, we said our theories postulate the existence of these properties/microstructures/entities in order to explain. From the fact that these entities do provide an explanation, and sometimes even a good explanation, we cannot infer that necessarily these entities/microstructures and so on indeed exist. In fact, we could also argue from a different perspective, if we like: we can say that, what we refer to as being the world, is itself a description of the world, is something that is made up by our theories. This certainly and first of all doesn’t sole the problem of correspondence, for there are at least 2 problems that come up here. First of all, getting a description of the world, if we follow this perspective, is precisely what we expect of our theory. This is what our theories are meant to do by classifying kinds as real kinds. We don’t start with the description of the world, we get it by our theories, we can’t assume that the world is itself already in description. Also, second problem here, we should consider that at least in principle, if we undertake this direction, there might be as many description as you can have theories of the world. So, equally (let’s again reason in principle) how many relations of correspondence can we think of? Equally, we should have as many correspondence as we have theories that describe the world in a particular way. And if there is more than one correspondence, how does this relation of correspondence provide with a strong criterion of access to the world? Its function as. Being what gives us access in a kind of secure and guaranteed way, seems to get downplayed. Now, note out, we should take into account that classifications and taxonomies of natural kinds are to some extend man-made, that they depend on our theory, to what theories produce. Some might even say that, because of this, there is really no other way that we can reach nature as it really is. The only way for us to have access to reality is by these classifications that are made by us, and that are made in such a way that they are meant to capture what’s out there. We always approach nature via some point of view/perspective/conjectural constructs, and all this is put forward by our theories (either scientific or common sense theories). We know that classifications and taxonomies are always put forward by virtue of some interest, of some goal, we classify in such a way because are goal is such and such, and we have a particular purpose, for example in science we want to explain things in a particular way and so on. Now, if we follow the discussion we had yesterday with Duprè, this is not per se an argument/ attack against realism/these metaphysical assumptions that there is a way that the world is, independently of us. That’s why promiscuous realism goes in this direction. Still, the temptation is to say that, if what we take to be the world is ultimately just a description of our theories, and therefore, if that’s the case, there’s no metaphysics involved here, just a matter of how we epistemologically reconstruct things about the world. This, indeed, is a temptation which is quite strong. There is a set of positions, in philosophy of science but not only, that goes under the label of anti-realism positions, for they embrace one form or another of this idea. It’s a range of perspectives that claim at different strengths of argument and with different consequences, depending on how strong my argument is, that what makes a kind natural depends on facts about us more than on facts that depend on the world. So, anti-realitist perspectives are said, you know, there are many different ways of being an anti- realist. One well known perspective goes under the label of conventionalism. Another is called “constructivism”, and that’s what we’re going to discuss today. What do we mean by constructivism? Constructivism, which is perhaps more widely referred to as social constructivism refers more or less to any perspective of a whole that what counts as a matter of fact is substantively determined by factors that have nothing to do with a mind- independent world. Normally, these are considered factors which are dependent entirely on us, on our history/actions/theories/ways of viewing reality. That’s why they are called by large “social”. By making social factors a kind of intrinsic and substantive determinant of what counts as being real or not real, true or not true, especially in science, social constructivism stands clearly on the opposite side of realism, why? For realism, and the main assumption of a realist position, is to say that scientific theories provide knowledge, but about a reality that is mind-independent, that is independent on us. Now, as a corollary of the constructivist assumptions, the notions such as proofs, references, ontology, they all become relative to particular contexts, contexts that are historical but also epistemic. And, they mean nothing outside this context, so there is not, in the possible sense of this position, a world out there and then different possible descriptions of this world. You have just different contexts from where you can actually argue that the world is one way rather than another, but independently of how the world actually is. So, why constructed? Why this word? What does this word make us reflect of? Constructivism comes from construction (lat. “construere”, namely to make or form), that carries with it an idea of building/putting together/assembling. But if we go back to the dictionary, we’ll see that under construction we have at least 2 meanings: we mean the activity, namely the way in which our thing is constructed, and the actual thing that is constructed. And this is a rather useful distinction, for it makes us reflect on this further distinction, which is theory related to this one, of construction as being a process, something that happens in time, and construction as a product, the end product of this process. So, construction occurs in time, which means it is accomplished by following a series of stages/ steps for combining/arranging things/parts in a particular way, whereas construction taken as the end product of this process, is the kind of thing that is itself pieced together in time, what we get at the end of this process. Now, this connection between process and product, which might appear rather harmless to us, in the hands of a constructivist becomes quite dangerous. It has possible important consequences for it leads us to a rather radical conclusion. Why? For, by putting these 2 things together (process and product) a constructivist intends to argue that the product so produced is not really inevitable. In one sense, we’ll change the process and the product will also change. Sometimes, it might even disappear, all together. There is no necessity here involved. We can see, and if we apply this thought to a scientific reference, we might say “well, you know, it’s because the way in which physics has developed that we ended up with things like electron: change the physics, and the electrons might well disappear”. Quite a radical conclusion here, quite worrying, in a sense. What is the product in question? When we say, here is the question, that “X is constructed”, so that is has a history of building behind it, what is the “X” we’re actually referring to? Let’s try to figure out what’s the argument behind this, what allows us to say that X is constructed and in what sense we say it. In order to do this, we’re going to use a mixture of these 2 authors: Ian Hacking and Andrew Kukla. Let’s take a look at what an argument for constructivism might look like, and then we’ll try to sort of comment and see where it leads us. We start from a first premise: X, or X as it is at present, is not determined by the nature of things. That’s the first step. Second step: X is the product of intentional human activity. Third premise: human activity, although necessary for the existence of X, is not itself necessary Therefore, conclusion: X need not have existed, or need not be at all as it is. So, if X is constructed, X is what it is because X is the product not of an objectively existing state of affairs, not because of any given order of things, but because of some kind of a contingent activity. X is not determined by the nature of things. This is what we’re told in the first premise. Now, the first premise, although it is a necessary condition for arguing/explaining what X is constructed, is not though sufficient by itself. It is necessary for the argument to come, but it is not sufficient: why? Because what we need to specify is that for our X to be not inevitable in our
argument/conclusion, it must be possible that, say, a scientist can make other choices, and these choices are reflected in a kind of world view that is consistent with this choice. In other words, the non-inevitability of my X is due to the fact that (premise 2) X is the product of intentional human activity, not any old contingent fact comes (?), but it has to be related to human activity and a kind of intentionality attached to it in the sense that, you know, it can be one way but it can also be another way. It is our theory/action/inquiries/history which makes X what it is. But, here is the important aspect, the theories/action/inquiries/history that make X what it is, although they are necessary for the existence of X (without them, X would not exist), they are not themselves necessary, for as we said before they could be otherwise, there are different ways of assembling theories or bringing history together. So, ultimately, the non-inevitability of intentional human action is responsible for the non-inevitability of X, so X need not have existed, or need not be at all as it is. We can see, it is not inevitable that there are electrons, for there are somethings in nature that make them what they are, and we refer to electrons because there are electrons. Okay, that’s the kind of a general scuffle for the constructivist argument. Then, the same argument can be read at different levels of strength, and it can be interpreted consequently in different ways. It can be read at the metaphysical level, which is the level of what is actually in the world, the level of the facts of the world we live in, which takes facts in a kind of whole encompassing way, not just the fact that we observe. Then, there is the epistemological level, which is the level of what we can know about those facts, the level of our theories/explanations and so on. And then there is the semantic level as well, the level of what can be said about these facts and about what we know about them and the language that we use to express this. Now, of course, we will realize that the strongest interpretation is the first one: to say that X is constructed, is to say that the very reality of X depends on human actions, or on the non- inevitable history of humans. We really need to reflect a little bit further on what it is this X that we’re referring to, precisely because it can be taken at all these different levels of interpretation. Here we go back to Hacking, for he presents us with a kind of almost simplistic kind of distinction, but it is use useful here to figure out what is going on. So, when we say that X is constructed, we can either refer to the so-called object, which is something, so to speak, in the world or the idea of X, so a conception/belief/theory/classification/ kind. Okay, so let’s take the strong interpretation of this distinction. We can say that X is the object, our constructed X is the object. In the end, there is no X as an object, there is just X as nothing but the idea that we construct of this object (there are no electrons, we just have a construction/idea of them) and the consequence here is to say “well, reality is constructed”. But, says Hacking, there is also a weaker interpretation. We can say that our constructed X is just the idea, and it is the idea that is constructed, not the object. Why? For in this case we admit that the object is in the world, so to speak (we’re talking in a very simplistic manner), independently of us, and what we construct is the idea of this object. So, back to the example of electrons. Now, notice that in both cases our X is an idea, but what it means to be and idea is quite different in the 2 cases. Why? For in the first case X is an idea because there is ultimately no object, so all we have is our constructed idea, whereas in the second interpretation X is the idea for the object exists in the world and the world cannot be constructed, only the idea of the world can be constructed. So, what we construct are our beliefs/theories/classifications/kinds and so on. We’ve seen the 2 interpretations: the constructs are ideas, in one case there is no object, in the second case it is an idea for the object is not the idea. In the first interpretation, we’ll see how we are actually getting quite in trouble, for from saying that what we take to be real is natural fact only a construction, back to this distinction that we just made, so it is just an idea of what is real, then we might go all the way and come to say that reality itself is a fiction, and socially constructed fictions cannot be objected. To the whole idea of objectivity of science goes out through the window. But, the first interpretation might also have one yet (?), that is it makes us reflect on the fact that the world of objects, in particular the world of objects we are referring to in science, is not simply given to us, so it is not just there, ready to be discovered as such. The objects that are questioned by science, from the most observable ones all the way to the unobservable, from what we see around us to molecules, to electrons, are always at least partly constructed. They are at least partly constructed by our theories/experiments/views, of course constructed in the sense that they need to be qualified in this context, for in this context, to say that they are not real, itdoesn’t mean that they are not real for there is no reality, that they’re not real for they’re constructed, that
they’re not real for there is no reality, but we’re saying that what we’re doing is constructing reality via the way in which our theories allow us to see reality to be. Now, these are subtle distinctions which need to be done for often, if we look at these debates about constructivist and anti-constructivist, which are quite ruthless/conflictual/confrontational, these subtleties often are neglected. So, in these debates, normally the way in which things are presented is that scientific objects or classifications are either discovered, for they are already there, and if they’re discovered then we’re a realist/anti-constructivist, or they’re totally invented, and in that we we are a constructivist. There is no middle way. So, kinds are either discovered or they are invented. Again, Hacking is more prudent in assessing the situation. So, he clarifies first of all that we should clarify and avoid the confusion between objects (something “in the world”) and ideas (a conception, a belief, a theory, a classification or a kind) in constructivist argument. When we talk about X, we need to make this distinction, or something like a distinction of this sort. And, the fact that claiming that scientific objects/kinds are at the same time or could be at the same time invented and discovered is not necessarily a contradiction in terms. Especially when we come to science, what we refer to in terms of the reference of scientific inquiries, what theories are actually about, they are partly constructed by our theories, and partly describe what is actually out there. One possible attempt to rebalance this divide between real and constructed has been put forward by yet another participant in this debate. She is an historian of scientific idea called Lorraine Daston. In her book “Biographies of Scientific Objects” (1999) she supports a view that she calls “applied metaphysics”. What is this position about? Just to summarize it, she says that the world of science, what we think of in terms of the world of science, is not really a static reservoir of given objects/classifications, it is rather a dynamic world, a world that is populated by objects which come into be and equally fade away as reference of scientific inquire, and they come into being and fade away depending on how epistemic inquiries actually develop. So, objects become reference of science depending on the scientist’s interest, the lines of questioning that they follow, the techniques that they use, of how much what they think it is important to inquire about are embedded in research practice/programs and so on. But this, says Daston, doesn’t mean that scientific objects are entirely a creation of scientists, and therefore they are not real/objective. The step that Daston instead takes here is to claim that, in an applied metaphysics, what she calls applied metaphysics, reality is a matter of degrees. What does she mean? Phenomena which are undoubtedly real in the colloquial sense, that they exist, can become more or less real depending on how much/deeply they are embedded in scientific practice. Take the example of the so-called preternatural philosophy: the latter inquires about preternatural objects (lat. “praeter natura”), so objects that are beyond nature, that are natural but yet they are irregular, they are deviations from the normal, they are extra-ordinary. And she has a whole list of these objects: images found in agates or marbles, comets presaging the deaths of kinds, a medusas head found in an hence egg in Bourdeaux, the power of flaxseeds to inspire prophetic dreams and so on. All these scientific objects came into be as scientific objects by means of preternatural philosophy around the 16th-17th century, and then equally they faded away when preternatural philosophy was supplanted by natural philosophy (Galileo and so on) between the 17th and the 18th century. So, watt’s the point here of discussing preternatural philosophy? Well, when scientific interest abandoned these objects, agates and marbles and flaxseeds and so on stayed exactly where they were, they only ceased to be scientific objects. In that sense, we can think of this dynamic between invention and discover, creation and reality. It might all sound a bit arbitrary this way of reasoning, this idea of coming and going of objects depending ultimately by scientific fashion. Also, it has been pointed out to Daston that her paradigm of object is sort of the everyday object, marble, seeds and so on, but what about scientific objects like electrons and so on? But she is making a more general point here, and she thinks of the object of science as being a rather uniform category of objects, so she’s rather catholic in her approach. She says that all source of objects belong to the same metaphysic which is applied, and by all source of objects, she means all of them. Here again, her list: dreams, atoms, monsters, culture, mortality, centers of gravity, values, cito-plasmatic particles, tuberculosis and so on, they are all equally belonging to the same metaphysics. There is no difference in that, you know, they are partly rthere/real, but
equally and for an equally important part for their very existence, they are partly invented and created and described and redescribed by our theory. By saying this, though, probably by approaching things in this way we don’t actually take into account a particular aspect to this debate within constructivism, which has to do with the fact that, as a matter of fact in this debate, objects that belong to the so called social science have been treated more harshly, and in a sense the arguments about construction, invention and discovery applied to the natural sciences and applied social sciences are not exactly the same. The emphasis on construction is a bit different in the 2 cases. With a perspective like Daston’s, all scientific object are put in the same pot, they’re all partly invented/discovered. But we’re entitled to ask: is there any difference really? At least between objects that belong to the so-called natural world and the objects that belong to the social world? Or should we just bypass the difference because of the reasons we have given? We aid, in fact, that in the way in which the debate about constructivism and realism has developed, people have not treated social facts and natural facts equally. So, on one side there was an almost a kind of dismissive attitude towards the social science object. Of course they’re socially constructed, what else? On the other hand, there was kind of a defense of natural objects, defense from being treated in a strict constructivist manner, for of course this goes against our testing intuition. Let’s see how this distinction might actually be relevant for us, and it might be interesting to pursue what the differences are between objects that belongs to a supposedly natural ontology and objects that belong to the social ontology. Here is quotation about socially constructed social objects from Kukla: “It’s almost universally believed that certain social facts (facts about social institutions, languages, social classes, governments, legal systems, economic systems and kinship systems) are what they are by virtue of our own actions, beliefs and intentions.” This doesn’t seem to be under discussion, in this passage. But the story, clearly, must be different when we come to natural facts. It is not easy to claim that the facts of nature, the facts that Science refers to, are human property. There are indeed a number of reasons why this idea of constructed nature, perhaps, appears to be harder to accept that the idea of constructed social facts. What are these reasons? First of all, the role of human agency in the case of social facts appears to be more obvious and better defined, even more crucial perhaps than in the case of natural facts. Secondly, to say that the existence of social facts is due to social factors seems to be less controversial than to say that the existence of natural facts is due to social factors. Thirdly, it appears almost counterintuitive that in the case of natural facts, no relevant distinction can be made between what belongs to a man-made reality (theories, classifications, actions…) and what belongs to an order of reality that is independent of human intervention. Overall, what seems to be harder to accept is that a constructivist ontology applied to scientific natural facts would ultimately get rid of an idea of reality that is objectively existing, external, an independent world of natural objects, processes that are real in nature. So, getting rid of a world that should be the most plausible referent of scientific find. And, also, it is equally harder to accept conclusions of this sort in the case of natural science for it would put us in the position of denying that we ever made discoveries. What we call discoveries are just subjective kinds of acquisition. Now, Daston positions herself in a kind of an autonomous way vis-à-vis these debates. She claims that all categories of scientific objects are partly real and partly invented, but the interesting question here is to see whether they are all partly invented and partly real in exactly the same way. And probably they aren’t. So we need to make some distinctions, and to do this we go back to Hacking. Hacking says that there is a big difference between quarks, on the one side, and children TV viewers/women refugees/criminals. What is the difference? How does Hacking portray these things? Let’s start from quarks: quarks are not aware, according to Hacking. A few of them may be affected by what people do to them in accelerators. Our knowledge about quarks affect quarks, but not because they become aware of what we know, and act accordingly. The classification “quark” is indifferent, in the sense that calling a quark a quark makes no difference to the quark. So, natural sciences are about indifferent kinds, and indifference kinds are somehow stationary targets, according to Hacking, which doesn’t mean passive, for many of these kinds to rather extraordinary things (think at what happens inside accelerators with these objects during experiments).
The point is that they don’t do any of the things that they do because they know what we’re doing to them. They are not aware in the way in which we classify/treat them. Now, the situation appears rather different when we come to the social domain. What happens there? Here’s Hacking description: “People are agents, they act, and they act, as the philosophers say, always under descriptions. The courses of action they choose, and indeed their ways of being, are by no means independent of the available descriptions under which they may act. Likewise, we experience ourselves in the world as being persons of various kinds. Such kinds (of people and their behavior) are interactive kinds. So, the reference targeted by social scientific inquiries, or targeted within the social domain, are so to speak always on the move. When people become the object of study of a social science, they end up interacting with the ways they are classified, and often they rethink themselves accordingly (“looping effect”). Think, for example, what happens during filed works in ethnographical studies: we have the ethnographers-anthropologists going to the field and interact with the individuals there, and often there have been reports of these individuals behaving in the way in which they think that the anthropologists want/expect them to behave. So, there is this active interaction with the classification and the people being so classified. Or, said in another way, there is a kind of feedback or “looping effect” involved when we classify people, and it is a looping effect where, according to Hacking, self-conscious knowledge plays much of a role. So, Hacking reminds us that the social sciences often are put under pressure in trying to prove themselves at being real sciences, just like the natural sciences, and therefore to produce real kind of people in the way which we have real natural kinds. But, easy Hacking, for this to be possible, the social sciences should endorse a certain epistemological picture that is not really adequate for the domain of inquire targeted by social science. This idea that there are objects to be searched for, the right kind, the kind that is true to nature, fixed targets, it just doesn’t work like that. This is not to deny, he argues to qualify, that the kinds social science deals with can also be in different subsets: think for example the case of child autism, the classification of child autism. We can claim that this is a social classification, but we cannot deny, equally, that the child/children in question can be the bearer of a particular type of pathology. So, an autistic child can be classified both interactively and indifferently. So, a social classification doesn’t exclude a natural classification, but even more so, this shows that the social ontology exclusively informed by indifferent classification doesn’t do its job properly, it does only part of the job. But this makes us reflect further on the role the so-called invention plays, in the case of this object/these objects. There is another philosopher of language who quite a few years ago published a book which was widely discussed, for precisely he targeted the way in which we construct social reality without being a constructivist. The philosopher in question is John R. Searle. What is his position here? He starts from this premise, which is a kind of ontological premise. For Searle, there is a sort of continuum between physical reality, logical reality, cognitive reality, all the way out to social or (the way he calls it) institutional reality. When we go from one extreme to the other, from the one side of the spectrum to the other, we come across phenomena that we can say owe their existence to the very fact that we represent them in a particular way. These are precisely what he calls social/institutional objects. So, why is this the case? How is it possible to claim that there are objects that owe their existence to the way in which represent them? In what way should we actually understand this claim? Let’s follow Searle’s line of reasoning. He says that there is a fundamental distinction we need to make here, between features of the world, which exist again independently of us (mountains, molecules…) and features which owe their existence to us, to our beliefs, the way in which we live (for example, money and marriage). Let’s take the example of marriage: the marriage does not exist outside the practice of people getting married. Marriage must mean something for the people who are contracting it (I would not get married if I don’t understand what getting married means). Marriage is an institution which comes about at a certain point in human history, it didn’t always exist, and again, to use the language we used at the beginning, it is not inevitable: it comes into be with certain human beings living in communities of a particular type, there are different ways of getting married and so on. Now, when we refer to external reality, we refer to both kinds of features, both the intrinsic, intentional and institutional features, and the extrinsic facts, of to use Searle’s own terminology, we include both so called brute facts and institutional facts. So, in order to have marriage, we
need to have real people and that particular contract that goes under that institutional name of marriage. Take a more immediate example, the one of the screwdriver: well, a screwdriver is a physical object which possesses some intrinsic feature. In this case, is is a solid piece of brass, with plastic handle. So, there is no doubt that the screwdriver is such a physical object. However, the intrinsic features of this physical object are not sufficient to classify this object as a screwdriver. It becomes a screwdriver for we use it in that way. In this way, a screwdriver exists as part of a world where we also find molecules, mountains and so on, and yet, we can say, it contributes a further category of identification for this object, which is, again in this case, people relative, relative to us, to what we believe we can put tase particular objects to do. Now, social or institutional facts, says Searl, behave in the same way, just like screwdriver. They exist because, first of all, humans had the capacity to impose status functional functions to objects. And these functions are never intrinsic to the very physics of the object. We intentionally assign these functions to particular objects to work in a particular way for us. Secondly, these functions are imposed to the categories of objects by a collectively sustained agreement or acceptance. There must be a community of people who agree that particular thing will be used for that particular purpose. Even more so, take the example of money: a banknote is a physical piece of paper, but in order to be considered as money, there need to be a community that works in a particular way and attributes that particular value and that particular function to that piece of paper. Without collective acknowledgment, that piece of paper is just a piece of paper, we would not call it money. This practice by which functions are imposed on the intrinsic features of a human-independent ontology has, according to Searle, an underlying logic that can be summarized in a little formula: X counts as Y in C. So, something physical/belonging (X) to the brute world counts as that particular thing (Y) in a context (C). This is the formula that expresses the constitutive rule that creates the very possibility for social objects to exist. Let’s see how Searle describes this formula and how it works. The application of the constitutive rule introduces the following feature: the term Y has to assign a new status that the object does not already have just in virtue of satisfying the X term (the X is a piece of paper, it is not money yet, Y is money); and there has to be collective agreement, or at least acceptance, both in the imposition of that status on the staff referred to by the X term and about the function that goes with that status. Also, there must be continued collective acceptance or recognition of the validity of the assigned function (I can’t just decide one day that this piece of paper is banknote and then I wake up the day after and say that I was just kidding: there must be a sustained contextual acceptance that that thing is actually that thing), otherwise the function cannot be successfully performed. What are we told here? First of all, that the move from X to Y creates classes of entities that we didn’t have before (money, property, political offices and so on), and these are all objects that do not exist outside the practices/contexts wherein these things take shape. As Searle puts it, in the case of institutional reality, the practice of attaching a sense/function to an object according to the constitutive rules creates the very category of potential referents. We create it. Again, if we want to say it in terms of formula, we would say that X counts as Y, iff (if and only if) C, all in the context of that practice. However (here’s the second thing we’re told), the entities that we have created this way are intelligible in the context of relevant practices. In the case of money, we need a referent practice of selling, buying, owning, earning and so on, as occurring in different societies. Again, if we want to say this in the terms of our formula, we say that X exists as Y iff we have this practice, namely iff X counts as Y in C. Third and last thing, the practice in question exists itself on the assumption that we have an X that we can apply to. Again, Y exists in C (money exists in a particular type of society which acknowledges it) iff (I have an) X. In other words, there is more to social reality than only social reality. Social reality is part of a physical environment and social facts exist on top of brute facts. So, this is a perspective that Searle defines as being “external realism”: he doesn’t see his position as being constructivism, and he is actually quite critical of constructivism. What does he think external realism amounts? He says that it is not an empirical theory, a theory of knowledge/language/truth, it is just an assumption that we make, it is the view/assumption that there is a way that things are that is logically independent of all human representation. Realism, in
the sense in which Searle takes it, does not say how things are, but only that there is a way that they are. Realism, as we said, works as a background presupposition. Only if we presuppose some kind of reality we are able to communicate with each other, we are able to function normally in our life. It has to be a presupposed assumption in all our discourses/actions. So, this is how we can summarize this type of argument: realism is a background presupposition (this is what we need to start from), if realism is this kind of presupposition then it is necessary (that’s what we start from), and if it is necessary, we can start build/figuring out the relationship between physical realism and social realism (by sharing the same background presupposition of external realism). This is called the transcendental type of argument in the literature, we go from the necessity to the possibility of a derivation. Now, notice, again, that there is a difference between scientific realism as we know it in science and this type of realism: why? Precisely for the reason that we’ve said: we’re not saying here what the unobservable features of the world actually are, we are just saying that they are, that they must be in order to have a world at all, or the world we all refer to when we communicate and interact. That’s the way the argument develops. As we said, this is a curious way of describing realism, but it is even more curious to see how Searle describes constructivism. The way in which Searle describes constructivism is such that he wants to take his distance from this type of perspective, he doesn’t want to be taken to be a constructivist. Constructivism is described by Searle as the view according to which we make the world, reality itself is but a social construct, alterable at will and subject at future changes as we see fit. So, this is not the kind of construction that Searle has in mind by putting us in front of his little formula about the construction of social reality. Well, again, we can criticize this view for we know that there is more to constructivism than the way in which Searle describes it, and in fact, back in those days, there was a big discussion between him and Hacking, and Hacking was criticizing Searle for taking the wrong view about constructivism, he says among other things that, in fact, the type of constructivism that the type of constructivism the Searle is portraying is a fail/stereotype, for after all, according to Hacking, constructivism might just be happy to hold that there is nothing to be said about reality of facts except what we identify as reality of facts, so it’s not the right way of putting it, for as we said before, reality might still be part of the picture in a constructivist outlook. It depends on how we approach it and what we intend to say about it. One final thought here: does constructivism have any merits? For a long time and still nowadays, constructivism is kind of harshly criticized, but is there any lesson/moral we can draw from looking at the constructivist argument in a non-banal way, taking the time of growing into the arguments and see what they actually mean and head for. Well, first of all, what we call reality is not necessarily a given, certainly, what we call reality and how we refer to reality in science and by our theories/classifications and so on. In this sense, even the world described by science, we can argue that is is partly constructed. So, we may good of what our theories tell us about this world, and this goes into the way in which we construct and describe our reference. So, claiming this is not necessarily an anti-realism. Reality can still be part of the picture, but we have a kind of well-defined/clarified and nuanced way of approaching.