Necessary Metaphysics

Interview by Richard Marshall.

Tuomas E Tahkois a metaphysician from the world's north who thinks about the contemporary Aristotelian tradition, the relationship between metaphysics and science, whether Kant was wrong, naturalised metaphysics, E.J. Lowe, natural kinds and essences, the law of contradiction, infinite regress, whether laws of nature are universal or not and why drawing continuous inspiration from the history of philosophy is a good thing. Pour yourself another phone and enjoy...

3:AM:What made you become a philosopher?

Tuomas Tahko:I have been interested in philosophical questions from an early age, but it was perhaps from age 15 onwards, in high school, that I began to gain the necessary knowledge and tools to formulate them. I had the good fortune of having a history teacher who had studied philosophy and covered the likes of Rousseau, Locke, Voltaire, and Montesquieu in great detail. In fact, in Finland, we also have some compulsory philosophy classes in high school and it was this same teacher, Ari Anteroinen, who taught me my first philosophy classes. Although I was initially very attracted to political philosophy and Rousseau's thinking in particular, it soon became apparent that metaphysics and the theory of knowledge tackle the questions that I was most deeply puzzled by. I was introduced to these questions especially through the ideas of Locke, Berkeley, and Hume, but even Kant was covered on my high school philosophy classes.

I went to study theoretical philosophy at the University of Helsinki without any detours, although I was aware about the lack of career prospects in this area, so I also studied some computer science and worked in that area during my studies. I did my MA (which in Finland automatically included a BA at that time) quite quickly, in three years, but although I enjoyed the courses, I discovered that no one in Helsinki was engaging with metaphysics in quite the way I would've liked. Towards the end of my degree, I discovered the work of E.J. Loweand read his The Possibility of Metaphysics(OUP, 1998), which influenced me deeply -- so deeply that I reached out to Lowe with some naïve questions regarding my MA thesis about essence and modality. He was extremely supportive and soon after in autumn 2005 I found myself in Durham working on my PhD in Lowe's supervision. I completed my PhD thesis, The Necessity of Metaphysics, in 2008. It's clear that Jonathan's influence on my work has been enormous.

3:AM:You’re an Aristotelian realist metaphysician aren’t you? This is a position that challenges the primacy of science to reveal what is real doesn’t it? Can you sketch the basic assumption of this metaphysical project?

TT:I do identify with the Aristotelian traditionand it was, again, E.J. Lowe who indoctrinated me to this tradition. But like him, I don't buy the picture wholesale. For instance, I don't accept hylomorphism. Sometimes I say that I'm an Aristotelian in a methodological sense rather than in terms of content. It's true that many Aristotelians seem to be committed to some kind of priority claim regarding metaphysics, i.e., it's the first philosophy and hence prior to science. But I think that the picture is somewhat more complicated than this simple priority claim might seem to suggest. For one thing, Aristotle himself apparently gave credit to empirical science and in some regards was also the first empiricist. I consider the relationship between metaphysics and science to be very subtle: there is a sense in which metaphysics is conceptually prior to science and deals with more general questions, but I don't think that it can get very far without empirical input. Sometimes the picture appears to be reversed as well: we gain some surprising empirical information, but it is only after this information is properly interpreted and analysed that we can truly understand it -- this process of interpretation is often partly philosophical.

In sum, although I am sympathetic to the tradition that is now called Aristotelian, I don't think it's quite clear-cut what the tradition amounts to when it comes to the priority question, for instance. For the same reason, I typically try to distance myself from the historical Aristotle and stick to what I know, namely, the contemporary movement that draws from Aristotle, but may or may not be historically accurate.

3:AM:Are you challenging the idea that philosophy is only good for clearing out conceptual muddles or helping reframe puzzles for investigation by the sciences, or any deflationary approach to metaphysics, such as the Quinean one which makes metaphysics continuous with science in its methods and aims ?

TT:Yes, I would challenge this. While I think that philosophy can help clear our conceptual muddles, I don't think that this is all it does -- this goes back to the issue regarding interpretation of scientific results mentioned above. As it happens, I've been recently working on this with Matteo Morganti. In our view, metaphysics and science both examine and explain reality, even though the means by which they do so differ. So metaphysics and science share their subject matter, but not their methods. But this non-overlap of methods does not entail that science and metaphysics are two completely independent ways of asking questions about a common subject and providing answers to those questions. Instead, we insist that the methodological autonomy of metaphysics can be maintained even if metaphysicians also employ the empirical results of science -- as they should. But I should add that this methodological divide does not necessarily go neatly along the a priori / a posteriori divide, as I think that both of these forms of justification and knowledge are used in both disciplines.

3:AM:So was Kant wrong as well according to this position, in that he saw metaphysics as investigating categories of understanding whilst you are interested in categories of ontology?

TT:To the extent that Kant really did think that the categories of understanding are 'in the head', then yes, my approach would be in stark contrast with the Kantian approach. Like most Aristotelians, I would regard ontological categories to carve out genuine joints in reality. But I try to be careful about historical claims and it is my understanding that even this issue is not entirely uncontroversial, i.e., that Kant should be read in the idealistic fashion, whereby everything is filtered through the categories of (human) understanding.

3:AM:How do you answer those like Thomas Hofweber who argue that this metaphysics is based on esoteric and inaccessible methodology?

TT:We actually had a nice discussion with Thomas Hofweber on the foundations of metaphysics back in 2012, on Philosophy TV. But I would perhaps answer a little differently now. Hofweber argues that many central notions especially in the Aristotelian tradition, such as 'essence' and 'ground', are esoteric and inaccessible to all but those who are trained in the discipline. I now think that the most forceful reply may be that science as well would be esoteric and inaccessible on this criteria. After all, is the notion of a 'Higgs field' or 'quantum entanglement' any more accessible to non-experts than notions like 'essence' or 'ground'? Of course, the former can be given an accurate mathematical formulation, but the latter can also be given a formal representation in logic. Moreover, the hope is that a scientifically-informed metaphysician might even be able to make sense of the esoteric notions of science with the help of this philosophical machinery. So I think progress can be made to illuminate the seemingly esoteric notions of metaphysics and certain areas of science alike.

3:AM:And what do you say to those who call for a naturalized metaphysics? Isn’t there a concern that it strays too far from findings in contemporary science to track any sort of ontology?

TT:The call for naturalized metaphysics comes in a variety of different guises and strengths. The best known version is now probably the one espoused by Ladyman and Ross in *Every Thing Must Go* (OUP, 2007). To an extent, I agree with this call: I think that philosophers ought to take the latest results of contemporary science seriously and do their best to construct philosophical theories in such a way that they are informed by the science. But it is easy to take this requirement for naturalized metaphysics too far, insisting that metaphysical theories are only of value if they manage to unify scientific theories, or to make predictions. It's ironic that the theory of ontic structural realism, developed in the mentioned book by Ladyman and Ross, may not itself qualify as naturalistic metaphysics on this criteria. Indeed, I think that we must be more moderate. Metaphysical theories can be naturalistic also in the sense that they are used to interpret scientific theories. This may also be understood to provide an *indirect* testing ground for metaphysical hypotheses, as I argue in joint work with Matteo Morganti. Interested readers might wish to read the final chapter of my An Introduction to Metametaphysics(CUP, 2015), which is dedicated to this question.

3:AM:Is the understanding of what is meant by ‘existence’ important to your approach? If the Aristotelian is working in a discourse considered as a collection of objects of thought – because it isn’t empirical – then isn’t this a form of Idealism? And why objects – why not just one big undifferentiated object – why have boundaries in reality written in as fundamental?

TT:As should be clear from my previous answers, I would insist that Aristotelian metaphysics -- at least insofar as I conceive of it -- is concerned with the very same reality as empirical science is. So it does not (only) study objects of thought, but objects of reality quite generally. Having said that, I do think that we also have some access to non-existent entities and to this extent it may be important to qualify what we mean by 'existence'. This does not commit philosophy to some esoteric form of idealism though, since science as well studies non-existent entities. For instance, we have been able to predict the chemical properties of various transuranic elements which do not occur naturally. Of course, we have then verified these predictions by synthesising these elements, hence making them existent rather than non-existent (albeit very briefly!), but this doesn't change the fact that we were able to study them before they existed. In a similar manner, we may be able to study merely possible objects, ones that could have existed, but never did (or will).

I should also mention that in principle there could be just one big undifferentiated object, perhaps in the lines of Jonathan Schaffer's priority monism, but so far it seems to me that the scientific and philosophical evidence supports the existence of more than one kind of object.

3:AM:So what do you think is required for fundamental ontology? There’s a lot of discussion around EJ Lowe’s claim that all we need are four fundamental categories. Some say this is too many, some too few. Bird thinks we can drop to three by dropping ‘natural kinds’. Heil thinks we can pare back to two. Simons thinks Aristotelians may require many more than four – maybe eighty-one! Could you sketch out what Lowe and the others are trying to do and where you stand on all this?

TT:Yes, this is certainly a core topic, although I must admit that it's one that I have tried to avoid in my own work, preferring to stay neutral. The reason for this is that I regard it extremely difficult to determine the fundamental categorical structure of reality. I am of course influenced by Lowe and his four-category ontology, but I think part of the problem is that here as well we should take heed from contemporary science, and the jury is still out there in this regard. But sooner or later I will have to commit myself. Since I regard Lowe's theory as one of the most sophisticated, it might be a natural starting point. I certainly stand with Lowe, Bird, and many others in that we need universals, even though I have some colleagues who vehemently argue in favour of a nominalist trope theory, dropping universals out of the picture. And I agree with Lowe in that we also need kind universals; this is because I think they do important work with regard to grounding the laws of nature. Bird, in joint work with Katherine Hawley, has attempted to develop a theory of natural kinds as complex universals, but I think this theory faces problems. One of them is that there may be natural kinds that are definable in terms of a single natural property -- a single property universal -- and hence would not constitute a complex universal (a possible case might be things like fermions and bosons, which may be distinguishable simply by their spin). But if we want to distinguish between natural properties and natural kinds, as I think we should, then we need some further account of natural kinds.

There is another recent complication, which draws on contemporary science: the already mentioned theory of structural realism. Now, I'm certainly not a convert as of yet, but I do take the theory seriously, and this has led me to wonder how much of the typical Aristotelian ontology could be salvaged if structural realism turned out to be true. It seems that what is arguably at the core of the Aristotelian theory of categories -- the category of substance -- would be at a risk, since structural realism puts pressure on the existence of individual objects. So what if we had to abandon the category of substance? We could perhaps still keep substantial universals (natural kinds) and we would also need properties and relations, given the work that they do even in structuralist accounts. But do we need tropes (modes)? I think that we might as well try to get by without them: if kinds can exist without substances, then properties and relations should analogously be able to exist without their postulated property/relation-instances. On this account, the universals would do all the work, and kinds would be 'bundles' of universals. But I should add that this is just something that I'm toying with at the moment!

3:AM:Kripke/Putnam set up a contemporary framework for natural kind essentialism but this has been subjected to a deflationary turn following various attacks. As an Aristotelian metaphysician you presumably take issue with deflationary moves, so how do you defend natural kind essentialism from its critics and deflationary saviours?

TT: We appropriately got into the topic of natural kinds already above. So it should be clear that I consider natural kinds to be part of fundamental ontology and hence would oppose the deflationary turn. That said, I also oppose the traditional Kripke/Putnam account of kinds, partly because I regard it as scientifically naïve. For one thing, it appears that it's much more difficult than is usually thought to determine whether something is a genuine natural kind, or simply a conventional classification. And it looks as if there might be fewer natural kinds than is often assumed -- here I actually agree with the deflationists. Biological species, for instance, seem to have boundaries so vague that I would hardly consider them as promising candidates for genuine natural kinds. As a side note, this is regardless of whether a Boyd-style homeostatic property cluster theory applies to them, because I think it's clear that the HPC-theory is actually not a realist theory about natural kinds. Perhaps the best we can do with regard to biological species is to identify 'living being' as a natural kind.

One complication here is that traditional natural kind essentialism defines kinds in terms of their intrinsic, microstructural properties. But this may be too strict: relational and macroscopic properties have to be taken into account as well. Quite generally, I think that claims for natural kindhood have to be assessed on a case by case basis. I have undertaken such a study with regard to the single most famous example, namely, 'water', concluding that the traditional microessentialist story about its status as a natural kind stands or falls with a more general principle concerning chemical substances. This principle states that there is a 1:1 correlation between the chemical properties of a chemical substance that constitutes a chemical kind and the microstructure of that substance. This gives us a criterion for chemical kinds that would appear to be defensible in many cases, perhaps including water.

3:AM:What does the law of non-contradiction look like from an Aristotelian point of view. Does this answer critics such as paraconsistent logicians like Priest who say that Aristotle’s defence of the law is feeble?

TT:My own view, which once again may or may not be faithful to the historical Aristotle, is that the law of non-contradiction (LNC) is a metaphysical principle that constrains reality, possibly in addition to our thinking about reality. Most paraconsistent logicians are typically concerned with a reading of LNC that entails only semantic dialetheism, i.e., only that there are true contradictions in language and perhaps in our thinking about the world. They may be right about this, I don't know -- language and thought can be rather vague, after all. But the step to metaphysical dialetheism, which suggests that there are true contradictions in reality, is a much stronger position. I have not seen any convincing examples in defence of metaphysical dialetheism; most of Priest's examples are quite easily defused as cases of semantic dialetheism. Quantum mechanics is the question mark here, of course, but the results are still open to interpretation to such an extent that I don't think they can be considered to constitute evidence one way or the other.

3:AM:Aren’t infinite regresses always sinister?

TT:This is a very hot topic at the moment -- Ricki Bliss and Graham Priest are editing a volume on the topic entitled *Reality and its Structure* (OUP, forthcoming) and there are other on-going projects looking into this. The question extends at least to metaphysics, epistemology, and argumentation theory, but I have only considered the metaphysical perspective myself. In metaphysics, infinite regresses have, until recently, been almost universally considered as sinister. This is especially true in the case of grounding / ontological dependence chains. I share some of the worries that have led people to think that such chains should terminate, but not all of the worries. A common worry is that composition could not get 'off the ground' if there is no foundation, because composite objects depend for their existence on their parts. But there is really no good argument in defence of this type of foundationalist intuition and anecdotal evidence suggests to me that it's not even an intuition that is so widely shared. For one thing, it seems to give a central position to the mereological (part-whole) hierarchy, but contemporary science tells us that the traditional mereological picture, which is commonly combined with mereological atomism, may be overly simplistic. All this leads me to think that we may have been too quick to dismiss infinite chains. A particularly intriguing case are chains that repeat the same structure over and over again -- Jonathan Schaffer has coined this as 'infinitely boring descent', but the case is anything but boring!

3:AM:Nancy Cartwright argues that laws of nature are not universal – they’re local, piecemeal and contextual. Is this a position your hybrid view of laws can accept?

TT:That's an interesting question. My own, 'hybrid view', to put it briefly, is that some laws are metaphysically necessary, but others (maybe even most of them) are metaphysically contingent. Cartwright's theory is a type of qualified anti-realism about laws, so in that regard it would be immediately opposed to my view, which is, after all, realist. But that may simplify matters a bit, since there is perhaps a more charitable way of reading Cartwright; while the fundamental laws are strictly speaking false, they might nevertheless be approximations. Well, this may already be against Cartwright's central ideas, but to compare our accounts in any kind of a fruitful manner will require some mediation. But here's one way to attempt a partial reconciliation: in my hybrid account, the contingent laws do not carry the same ontological weight as the necessary ones, because they are not grounded in natural kinds (whereas the necessary laws are). Perhaps the reason for this (although I would hesitate to commit to this) could be thought to be exactly because they are local, piecemeal and contextual -- strictly speaking only approximations, perhaps sometimes even merely instrumental. Now, I would put this down to the fact that we are still in the early days of finding out what the truly fundamental laws are and what they entail. The contingent laws could then be considered to be useful but inadequate results of our feeble attempts to capture the fundamental laws. With some effort, this could be seen to connect with Cartwright's idea about phenomenological laws vs. theoretical laws, i.e., the contingent laws could be interpreted as phenomenological, whereas the fundamental, metaphysically necessary laws would be theoretical. But it must be said that this is a rather idealized comparison, one shouldn't read too much into it! My own work is much more closely aligned with that of scientific essentialists like Bird and Ellis than with Cartwright's.

3:AM:Does your approach to metaphysics suggest we should be going back to the medievals as well as the Ancients to assess our views about the world and ourselves. Doesn’t your approach reverse the usual way we look at the situation: they were the dark ages and we were enlightened but isn’t your approach saying the Quineans and the naturalists the ones who are missing what’s actual?

TT:In general, I think that we should continuously draw inspiration from the history of philosophy, from all periods. I frequently consult my colleagues working in the history of philosophy and have learned many valuable lessons applicable to contemporary metaphysics from this. Aristotle, in particular, is an obvious inspiration for my own work. But given the parallelism between metaphysics and science that I have been underlining in this interview, it should be equally clear that I would hesitate to take historical views to be much more than an inspiration. That's because the ancients and medievals did not have the vast body of science to draw on that we do now. Of course, there may be some philosophical topics where this matters less, even within metaphysics. But here I think Aristotle might very well agree: he arguably did take the science of his time quite seriously and I'm sure he would've been thrilled by the advances in our pursuit for truth that science has enabled. So as a methodological precursor, I regard Aristotle and indeed many other historical figures to be very much worth drawing on.

3:AM:And for the readers here at 3:AM, are there five books, other than your own, that you can recommend to take us further into your philosophical world?

TT:Sure:
E.J. Lowe, The Possibility of Metaphysics(OUP, 1998)
Alexander Bird, Nature's Metaphysics(OUP, 2007)
Matteo Morganti, Combining Science and Metaphysics: Contemporary Physics, Conceptual Revision and Common Sense(Palgrave Macmillan, 2013)
Kathrin Koslicki, The Structure of Objects(OUP, 2008)
David S. Oderberg, Real Essentialism(Routledge, 2007)

ABOUT THE INTERVIEWER
Richard Marshallis still biding his time.

Buy his book hereto keep him biding!