:: Article

The Fragmentation of Being

Interview by Richard Marshall.

Kris McDaniel works mainly in metaphysics, though he has research interests in ethics and the history of philosophy. Here he discusses Heidegger and his analogy of Being and Beings of Reason, holes, Heidegger’s distinction between present-at-hand and ready-to-hand objects, Van Inwagen on Heidegger’s understanding of Being, whether existence is a property, whether Being can come in degrees, ontological pluralism, and ontological monism, composition, why whole’s aren’t identical with their parts, carving nature at the joints, whether parts of reality are eternal, and whether we can decide whether a metaphysical system is right or not.

3:AM:  What made you become a philosopher?

Kris McDaniel: The short answer is wonderful teachers! I attended Western Washington University in Bellingham, Washington, a smallish public university. I think I was sort of lost when I went to college. I didn’t really have any sort of idea about what I was going to do with my life. I enjoyed playing music, and I entertained the idea of pursuing some sort of musical career while holding down manual labor jobs. But that’s about as far as I got. I didn’t take my classes very seriously my freshman year.

During my second year in college, my girlfriend suggested that we take a bunch of classes together. One of them was Hud Hudson’s Introduction to Ethics class. It blew my mind. I was amazed that one could use logic to think carefully about issues in applied ethics. And I was really taken by the thought that the goal of philosophical ethics was to develop comprehensive, systematic, and true ethical theories. That class got me hooked. Hud was and continues to be a role model for me, and a great friend. I owe him a tremendous debt that I will never be able to fully repay.

For the next few years, I took as many philosophy courses as I could, read as much philosophy as I could, and engaged with my peers and teachers philosophically as much as they would permit. My teachers at Western were fantastic. In addition to Hud Hudson, I had classes with Frances Howard Snyder, Neil Feit, Tom Downing, Phil Montague, and Ned Markosian. I took many classes on metaphysics, the history of philosophy, ethics, and epistemology there. Many of my brothers and sisters from Western Washington have gone on to have successful careers in philosophy.

My luck continued when I went to graduate school, were I also had wonderful teachers, including Phil Bricker, Gary Matthews, Lynne Rudder Baker, Keven Klement, Jonathan Schaffer, Fred Feldman, Michael Jacovides, and many others. At the time I attended, Umass-Amherst was very strong in metaphysics and the history of philosophy, and my passion for both continued to grow, although my focus was certainly on metaphysics.

Although my teachers had very different philosophical perspectives and teaching styles, they all demonstrated a kind of seriousness and earnestness about philosophical inquiry that I think is what inspired so many others to follow in their paths. Philosophy was never a game for them. I’m nowhere near as good of a teacher as my mentors. But I hope to convey to my students this same sort of respect for philosophy. I’ve found that doing philosophy can be joyful, stressful, exciting, and sometimes even tedious or boring, but it I have never once felt that it wasn’t an important and worthwhile activity. Philosophers are the mountain climbers of the mind.

In general, I feel incredibly lucky. Things could have gone much differently for me, and my life probably would not have been anywhere near as good as it in fact is.

3:AM: Let’s start with some hard-core Heideggarian metaphysics. You’ve tried to work out what Heidegger thinks is the important connection between the analogy of being, truth and ontological idealism haven’t you. So first, can you sketch for us what the analogy of Being and Beings of Reason is in his work and where historically it comes from? And can you also update us on what we are to make of terms like Dasein, Being and the like as used in this context so we can follow what’s going on!

KM: That’s a big task! The idea that “being” is analogous or “said in many ways” is suggested by Aristotle. There are all sorts of entities: people, tables, lumps of matter, numbers, for example. Let’s agree that each of these and other things that are among what there is exist or have being. But do they exist or have being in the same way? If the way in which a person exists isn’t the same way in which a number exists, there are two ways to exist. But what makes a way a way of existence? Maybe there is a paradigmatic way of existence—perhaps it’s the way in which substances exist—and the other ways of existence deserve to be so-called because they are related in some way to how substances exist. (For example, an accident has a different kind of existence, existence in, and what an accident exists in is a substance.) In a nutshell, this is the analogy of being.

I think Heidegger, at least in the mid to late 1920s, believes in the analogy of being, or at the very least adopts it as working hypothesis because it is hard to see how to make headway on the question of the “meaning of being” unless the various ways of being are in some way related to one another. But he definitely rejects the claim that the way in which substances exist is the primary or central mode of being. So if the way substances exist isn’t the central mode of being, which mode of being is? Here is where Dasein enters into the picture. The working hypothesis is that the way Daseins that exist is central.z

Dasein is the kind of thing that includes you and me. I think “Dasein” basically refers to the same things as “human person”. That said, there’s some scholarly controversy over whether Dasein is a count noun or a mass noun, and some scholarly controversy over whether Daseins are individual human persons or whether each of us is something like a case of Dasein (similar to how individual colds are cases of the common cold).

Ok, back to the analogy of being. Why suspect that the way Dasein exists might be the central way of existence? Because we have some insight into Dasein’s way of existence, and it is constitutive of that way of existence that Daseins care about the fact that they exist, that they worry about their continued existence, that they fear their nonexistence, and so forth. As Heidegger more or less puts it, “Being is an issue for Dasein”. So part of what it is for a Dasein to be is that a Dasein cares about being. The other ways of existence that we are familiar with don’t seem to refer to being in this way. So if we are looking for a good candidate for a central mode of being, beginning with the one mode of being that has reference to being built into it seems like a good place to start!

3:AM:  How are we to even understand Heidegger’s idea of a ‘being of reason’ that denies, for example, that necessarily an object has a hole if and only if it is perforated and that requires that only if Daseins are can holes be said to be? Surely, if we have the perforated object then we have a hole regardless of whether we’re around? Or is Heidegger saying we literally bring something else into existence on top of the perforation?

KM: The idea of a being of reason isn’t original to Heidegger—it’s an idea that can be found in a number of medieval philosophers, and is arguably in Aristotle as well. I’m not even certain that it is in Heidegger, but I do think that there’s a reasonably strong indirect case to make that it is. (For those who might be interested, the case is discussed in my paper, “Heidegger and the “There is” of Being”.)

Here’s a rough and ready albeit contentious way to understand beings of reason. There are two kinds of ways of existence. The first kind consists of those genuine or fundamental ways of existence that are had by the real things. If we have an Aristotelian ontology, for example, then the real things are substances and the various kinds of accidents, and the genuine ways of existence are the categories these objects fall under. But in addition to these entities, we need to recognize further entities—and therefore we need to recognize another way of existence, namely the way in which these further entities exist. These further entities are things like shadows, holes, blindness, and other deprivations and absences, and maybe fictional entities like chimera or creatures of dreams or mythology. Why do we need to recognize these entities? Because there are innumerably many true sentences about them, and so they must exist in some way. For example, “You should be careful when driving, because there are three holes in the road” might be true, and it is about holes. But holes don’t fall under any of the categories—they aren’t substances or accidents. So there must be some other way they exist, which is a deficient or tenuous manner of existence.

So far, I approve of this line of thought. But there’s a further step taken that you are picking up on and that you are right to be concerned about. This further step is that beings of reason are mind-dependent entities! On the traditional way of thinking about beings of reason, an object could be perforated even though there is no hole in it. According to the traditional way of thinking, a necessary condition for a hole to exist is that there are finite beings who cognize the world in certain ways. Here is where I disagree. In fact, I think it’s a conceptual truth that if there are holes, then, necessarily, an object is perforated if and only if there is a hole in it.

One insight of the traditional way of thinking is that the reason that we talk and think about holes and other so-called beings of reason is that we are intellectually imperfect. Human beings would have an enormously difficult time navigating our environment if we didn’t talk and think about holes. (As an exercise, try to figure out what you’d say instead of “Look out for those three giant holes and those two smaller ones!” And now try to figure out how you’d systematically replace talk of holes from our discourse. Not an easy task, and I think probably not even a doable one.) In a sense, there are situations about which we have to reason by thinking in terms of holes—we have to posit things that aren’t genuinely real in order to cope with how real things are. But it doesn’t follow from this insight that holes are mind-dependent. If there is a hole in the center of our planet, it would have been there even if no finite minds had ever existed.

So I believe that we should recognize a deficient way of existence had by things like holes, but we shouldn’t think that holes are mind-dependent. For this reason, I avoid calling them beings of reason, since that suggests mind-dependence, and instead call them “beings by courtesy”.

Incidentally, this distinction between two kinds of existence—genuine reality vs. a deficient and possibly mind-dependent mode of existence— isn’t unique to the Aristotelian tradition. For example, something like this distinction also seems to be found in some of the Buddhist philosophers of the Abhidharma, who distinguish between conventional and ultimate existence. Consider a chariot and the simples (that is, entities without parts) that compose it. The simples are ultimately real. They are “really” there in that peculiar metaphysician’s sense of “really”. The chariot isn’t ultimately real, but it still has a lesser kind of reality: it is conventionally real.

Ok, let’s circle back to the philosophy of Heidegger around the time of Being and Time. Here are some puzzling data points. Heidegger denies that being is a being; Heidegger denies that ways of being are beings. Yet Heidegger sincerely and emphatically utters a lot of sentences about Being and its modes. Heidegger also says puzzling things like Being “is” only if there are Daseins, although beings could exist without any Dasein existing. (He uses something like scare-quotes when he says this too.) These statements typically appear in a larger context in which he is talking about the connections between being and truth. The analogies between how beings of reason were thought of and how Heidegger here speaks of being and its modes are striking.

3:AM:  Is Heidegger’s metaphysics of material objects in Being and Time best interpreted in accordance with the two domains view or the single one? Doesn’t the fact that we only use a single existential quantifier suggest that there is just a single realm, and everything that exists is in that?

KM: The one domain view and two domains view are interpretations of Heidegger’s distinction between present-at-hand and ready-to-hand objects. Roughly, present-at-hand objects are things like lumps of matter or fundamental particles, and ready-to-hand objects are useful things like tools, e.g., hammers, circular saws, nail guns, drills, miter saws. (I built a deck for my house during the summer and fall this year, so these examples really jump out for me.) The one domain view says, as the name suggests, that there is just one set of things here but two different ways of considering or encountering them. (One can view a hammer as a hammer or as a lump of matter.) The two domains view says, as the name also suggests, that there are two distinct sets of entities: no ready-to-hand entity is a present-at-hand entity. (So the hammer is not identical with the lump of matter that coincides with it.)

There are strong reasons to favor a two domains interpretation. Heidegger seems to ascribe different modal, temporal, and axiological properties to present-at-hand things and ready-to-hand things. He seems to explicitly say that they differ with respect to their essences and to how they are individuated. He seems to ascribe different ways of existence to them.

That we use a single existential quantifier—if in fact we do use only one—doesn’t by itself constitute a problem for the two domains view as an interpretation of Heidegger. At most, maybe that would provide some small reason in favor not accepting a two domains interpretation of our own metaphysics rather than Heidegger’s.

The question of whether Heidegger believes in ways of existence is not wholly independent of the question of whether the two domains interpretation is correct. But these questions are to some extent independent. Suppose there is only one way to exist. Still, a hammer and its coinciding lump of matter might be numerically distinct. (That’s a pretty popular combination of views in contemporary metaphysics.) Suppose there are many modes of existence. Still, a hammer might be identical with its coinciding lump of matter. Moreover, the claim that this hammer/lump of matter is one thing but enjoys two distinct ways of existence is coherent. In principle, there are four interpretive possibilities here.

That said, I do think that the two domains interpretation nicely fits with an interpretation of Heidegger as a believer in modes of being.

3:AM:  Van Inwagen accuses Heidegger of mistaking Being for nature. He says all Being is about is whether something exists or not, not whether it has a certain nature or properties. I guess I see his point; existing things can be vastly different and we can often hope to understand those vast differences – but surely existence isn’t one of them. Is he right in this?

KM: I think not. Van Inwagen’s view is plausible but certainly one can reject it, and, unsurprisingly, I think there are plausible reasons to reject it.

Interestingly, one reason to reject van Inwagen’s position might come from considering not objects that are qualitatively very dissimilar but rather from considering objects that are very much alike. Consider the difference between an actual object and a merely possible intrinsic duplicate of it. So, for example, consider yourself and a merely possible duplicate of yourself. They are intrinsic duplicates—there is no difference in their properties. (So, for example, if you are n cm tall and have a mass of m kg, your duplicate also is n cm tall and has a mass of m kg.) With respect to your essence or nature, you might also be on a par. But one of you is actual while the other merely possible. The fundamental difference between the two of you doesn’t seem to consist in a difference in qualities between you and your twin, but rather the difference is that you exist in different ways; in general, merely possible objects don’t exist in the same way that actual objects do.

3:AM: Do you think existence is a property, and if so, what kind?

KM: I’m unsure. My working hypothesis is that it is a higher-order property of properties or relations. This is a fairly standard view about existence. On this view, “cows exist” is equivalent to “the property of being a cow has at least one instance”.

But I think the arguments for the view that existence is a higher-order property are not terribly strong either. One popular argument has two premises. The first premise is that attributions of existence are a lot like attributions of number. You can say that the number of cows in the field is 20, and you can say that cows in the field exist. When you say the latter, you communicate that there is at least one cow in the field. The second premise is that numbers are properties of properties; in the example just mentioned, the number 20 is attributed to the property of being a cow in the field. The conclusion is that existence is also a property of properties. The main weakness of with this argument is that it isn’t clear why numbers couldn’t be properties of individuals. Sure, no individual cow in the field is 20 in number, but the cows collectively are 20 in number—and this seems to be an ascription of number to some things rather than to some property.

When I first started thinking about modes of being, it was important to make sense of ontological pluralism without taking being or existence to be properties, or even any sort of entity at all, since at least one well-known ontological pluralist—Heidegger—doesn’t think that being or modes of being are beings. (And even this claim about Heidegger is more complicated and contentious—recall the discussion of your third question!)

One of the current things that I am working on is whether existence could be a second-order property even if modes of existence are first-order properties. And one of the graduate students at Syracuse, Byron Simmons, is exploring arguments for taking modes of being to be first-order properties. I’m excited to see how this all develops.

3:AM:  You argue that Being can come in degrees don’t you? This seems mighty strange – if by Being you mean exist how do you defend this proposal?

KM: Not only are there different ways to exist, but some of these ways of existence are more metaphysically fundamental than others. And some objects have more existence than others: as I mentioned earlier, a hole in a piece of cheese is less real than the cheese itself. On the metaphysics I prefer, there aren’t two independent scales here, but rather an object’s degree of being is proportionate to how fundamental its most fundamental mode of being is.

Historically, the view that some things have more being than others was pretty popular. In response to earlier questions, I alluded to one reason for accepting this view: entities like shadows, holes, heaps, or piles can be referred to, can be causes or effects, have conditions under which they are identical or persist. (I have children, so I am very familiar with arbitrary heaps of things; every evening, after my children are in bed, my wife and I disassemble heaps consisting of clothing, paper, toys, family pets, children from down the block….) All these facts are reasons to think these entities exist in some way. But they aren’t obviously reasons to think that they exist in the same way as we do, or that they exist to the same extent that we do.

There are also theoretical advantages to be gained if being comes in grades or degrees. One of them is that it is a nice theory of metaphysical fundamentality. For example, on the view I like, the naturalness or fundamentality of a property just is how much being that property has. On this view, merely disjunctive properties, such as the property of being touched by Justin Bieber or having a mass of 135 kg—have less being than their disjuncts.

3:AM: You like ontological pluralism, the doctrine that some things exist in a different way from other things. Historically this was popular – how come it isn’t so popular now? Is it because logic got more powerful, and can you explain how we should understand your version of the doctrine – perhaps by looking at how you think we should think about holes or shadows? And how does analogy help your position?

KM: I don’t think the power of logic explained why ontological pluralism became less popular in analytic philosophy circles. I am not sure it became substantially less popular in “Continental philosophy” or in other traditions.

It’s hard to deny the impressive power of today’s logical systems. That said, I think that some of the pioneers of logic were ontological pluralists. There’s a nice paper by Ben Caplan called “Ontological Superpluralism” that discusses Russell’s and Frege’s versions of ontological pluralism. And ontological pluralism can be stated in as logically rigorous a way as ontological monism, using basically the same resources of modern quantification theory. Jason Turner has some nice papers on logic and ontological pluralism.

I don’t think that the neglect of ontological pluralism in analytic philosophy was due to particularly strong arguments against it. I don’t have a positive theory of why it was neglected! It’s worth keeping mind that a lot of metaphysical views were neglected by analytic philosophers for a good chunk of the 20th century. Metaphysics doesn’t start getting awesome again until sometime in the 1970s.

I just want to make ontological pluralism great again.

3:AM: Your position is contrasted with ontological monism of some stripe. You say that an ontological monist will be embarrassed by holes because she says all things ‘enjoy the same kind of reality enjoyed by fully fledged concrete entities such as ourselves.’ But why doesn’t a solution such as van Inwagen’s, who paraphrases holes away into talk of perforated objects, not convince you? ( After all, his approach seems to offer a much simpler ontological landscape and avoids the weirdness of degrees of existence which seems very odd.)

KM: What philosophers intend to do when they offer paraphrases is not always clear. A philosopher might offer a paraphrase of a sentence as a way of disavowing any objects whose existence is entailed by that sentence. I don’t have an objection with that project per se, but I am less interested in that philosopher’s disavowals than I am in the question of what the sentence actually says. And that’s because I take these sort of sentences to be true, rather than, for example, falsehoods that are convenient to utter in certain contexts.

I don’t think there’s any real hope of systematically paraphrasing away talk about holes. Consider a context in which “there are as many holes in the cheese as there are crackers on the plate” expresses a truth. Maybe a philosopher could say, “there is some number, n, such that n numbers the crackers on the plate and the lump of cheese is n-perforated.” Ok, but what about, “there are as many holes in the cheese as there are crackers on the plate, and some of them are bigger than the crackers”? And you know that there are even more complications that could be introduced. At the end of the day, I don’t see a good argument from linguistics for the claim that “there are as many holes in the cheese as there are crackers on the plate” has a different structure than “there are as many lions in the zoo as there are people at the gift store.” I don’t see a good argument from linguistics to think that the apparent truth-conditions of both sentences aren’t exactly what they seem to be.

I agree that holes are ontological degenerates. But I want to explain that fact by telling you something about holes, rather than by giving you a theory about “holes”. Holes are ontologically dubious but there is nothing linguistically peculiar about “holes”. What we need—what I offer—is a theory of what it is to be ontologically degenerate.

3:AM:  Why are considerations of the gradation of being and ontological pluralism important for thinking about the questions as to why there is something rather than nothing?

KM: The question of why there is something rather than nothing is about something and nothing. And if something and nothing come in grades and ways, then there are also the questions of why there is something that is to a certain extent rather than some other extent, and for each way, there is the question of why there is something in that way rather than nothing in that way. From a metaphysical perspective, these might be deeper questions to ask than the original question. (Sometimes philosophical investigation teaches us that the original question we began with wasn’t the best question to ask.)

3:AM: Another key area of metaphysical debate regarding ontology is the idea of composition. So where are you in this dog fight – a modest pluralist, a compositional nihilist or something else completely? Could you also say why composition is so important to many ontologists and the kinds of problems they’re having to face?

KM: The topic of composition is deeply connected with so many other topics in metaphysics. If you want to know what there is—a question that some philosophers think of as the central question of ontology— you’ll want to know when composition occurs. The topic of composition is connected with the topic of fundamentality: are parts in some way metaphysically prior to what they are part of? Are wholes in some way less real than their parts? (See Karen Bennett’s recent work for a conception in which composition and fundamentality are aspects of an underlying relation she calls building.) The topic of composition is connected with the topic of objectivity: to what extent do our classificatory practices result in the existence of objects that are thereby classified? The topic of composition is connected with the topic of vagueness. A theory of when composition occurs that we find plausible is probably going to be one that implies that it is sometimes vague whether composition occurs. But can it be vague whether something exists? The topic of composition is connected with the question of what sort of thing we are, and in general, with the structure of the world. Are material things just made out of lumps of matter or particles or what not, or do we also have formal parts that structure and arrange our material parts? The topic of composition is connected with the topics of formal logic and formal ontology. Is the relation of parthood like the relation of identity, the latter of which is a logical relation that topic-neutral, i.e., applicable to entities regardless of their essence or nature?

In more than one way, I am a modest pluralist about composition. First, I am a modest pluralist about modes of composition. There are different kinds of entities and corresponding to them are different relations that “build” objects into wholes of some kind. There are regions of spacetime, material objects, and facts about them. And for each kind of entity just mentioned, there is a distinct kind of parthood. So-called classical mereology governs the parthood relation enjoyed by regions of spacetime. In the system of classical mereology, parthood is a two-place relation that is transitive, composition is unrestricted (all groups of regions compose a larger region), and composition is extensional (if “two” regions have exactly the same parts, they are identical).

But classical mereology doesn’t govern the parthood relation enjoyed by material objects. That relation is a three-place relation (x is a part of y at R, where R is a region of spacetime), and unrestricted composition doesn’t hold. There’s nothing made out of all the noses in the world.

With respect to facts (such as the fact that you are interviewing me), both unrestricted composition and extensionality fail. Even if I interview you at the same time that you interview me, the fact I am interviewing you is a different fact than the fact that you are interviewing me. But these facts have the same components: you, me, and the relation of interviewing.

I’m also a modest pluralist in the sense that I think composition sometimes occurs and sometimes doesn’t. As just mentioned, I don’t think there’s anything made out of all the noses in the world. And in general, I don’t think there are compelling arguments for fully unrestricted composition, i.e., compositional universalism. But I’m not a nihilist either. There are people, dogs, trees, molecules, and cars. However, because I also think that there are degrees of being, saying this doesn’t settle whether people, dogs, trees, molecules, and cars all exist to the same extent. I’m much more confident that living things have a robust kind of being than I am that artifacts like cars do.

3:AM:  If a whole is numerically identical with its parts and wherever there are some things there is a whole composed of those things it looks to some as if there is some kind of logical dependence between the two positions. But you don’t think so. What’s at stake in this and what’s your argument for denying that composition as identity doesn’t entail universalism?

KM: At the end of the day, I don’t think a whole is identical with its parts, but I can see the considerations that favor the view. Here’s an example from Donald Baxter. A farmer has a farm that consists of three plots. The farmer sells each of these plots to three different people. He then sells the whole farm to a fourth person. The farmer is feeling pretty good right now—he is raking in the dough. He reasons that he didn’t do anything wrong. He owned the farm, so he had a right to sell it. He owned each of the plots, so it was ok to sell each of them too. And the farm isn’t identical to any of its plots, just as you aren’t identical to any of your proper parts. So the farmer had four things to sell, and sell four of them he did. What’s the problem?

Composition as identity has a nice explanation of what’s gone wrong. The farm is identical with its three plots. It’s not identical with each of them individually, but with all of them collectively. So when the farmer sold the three plots, he did sell the farm. He didn’t sell the whole farm to any single person, but he sold it to three different people. No one person now owns the farm; the three owners of the plot now collectively own the farm.

That’s a nice explanation of a puzzling phenomenon. If we reject composition as identity, we need a different explanation, and it is less obvious what it should be.

Philosophers love to find new arguments for positions that they already accept. So fans of compositional universalism would be pleased if compositional universalism were entailed by composition as identity. But philosophers also search for deeper explanations, and this is a different activity (on the face of it) than producing arguments. You might have a good argument for a conclusion, and so be convinced that it is true, without understanding why it is true. Composition as identity seems like it could explain why parthood has the features that many take it to have. Suppose that classical mereology is true across the board—that is, it is a theory that is true of everything without restriction, and it is the theory of the only parthood relation that there is. (In other words, suppose classical mereology and reject the kind of compositional pluralism I mentioned in response to a previous question.) As mentioned earlier, classical mereology consists of three axioms: the transitivity of parthood, the extensionality of composition, and compositional universalism. Each of these has been contested by philosophers, but I think I have listed them in order of least to most controversial. Wouldn’t it be neat if we could explain why these axioms are true rather than take them as, well, axiomatic?

Composition as identity seems to explain the transitivity of parthood and the extensionality of composition well. Identity between one thing and another is transitive. But this principle also seems true, at least if we allow that many things can be identical with one thing: if x is one of some things that are identical with y, and y is one of some things that are identical with z, then x is one of some things that are identical with z. And given composition as identity, v is a part of w if and only if v is one of the things that are identical with w. From these two claims, it follows that parthood is transitive. That’s neat! Composition as identity seems to explain why parthood is transitive. Now consider the extensionality of composition. If many things can be identical with one thing, the following principle is plausible: x is identical with some things and y is identical with those same things, then x is identical with y. Now, if composition isn’t extensional, there can be two distinct wholes composed of the same parts. But then our plausible principle would fail. So composition as identity seems to explain why the extensionality of composition holds.

All that is missing is compositional universalism. If composition as identity also entails compositional universalism, then the three axioms of classical mereology would all be explained by a single principle. Reducing three axioms to one would be real progress in metaphysics.

However, I don’t think that composition as identity entails compositional universalism. Composition as identity tells us that if there is a whole made out of some parts, then the whole is identical with those parts. But composition as identity doesn’t tell us that whenever there is some things, there is a whole composed of and identical with those things. (Ross Cameron nicely observed this.)

I offer a direct argument: there are three mutually consistent theories that jointly entail composition as identity and the denial of unrestricted composition. Since these three theories are jointly consistent, what they entail must be jointly consistent as well. And so composition as identity doesn’t entail unrestricted composition. The three theories are (i) an incredibly modest pluralism, namely that there are least two things, (ii) the view that properties and relations are identical when they apply to exactly the same things in any possible situation, and (iii) compositional nihilism, the radical view that some things xx compose something else y only when the xx are really just one thing, and that thing is y. (It is common in the mereology literature to understand “part” and “compose” in such a way that everything is a part of itself, and everything composes itself.) Given these three theories, the composition relation is numerically identical with the identity relation, and that is the strongest way to formulate composition as identity. Yet (i) and (iii) imply that compositional universalism is false.

As noted earlier, I reject unrestricted composition/compositional universalism, and composition as identity. It would be nice if there were a tight connection between the two views I reject, because then an argument against one of them would be the basis for an argument against the other one! In that sense, more would be at stake for me if composition as identity did entail universalism.

I think we probably should reject composition as identity. My main reason for rejecting composition as identity is that I think that composition as identity implies that genuinely emergent properties are impossible. You can think of a genuinely emergent property as a fundamental property of a whole whose exemplification isn’t determined by the properties or relations of the whole’s proper parts. The argument roughly is this: if a whole just is its parts, then if you duplicate all of the parts (and the relations between them), any other whole that is made out of those duplicated parts must be a duplicate of our original whole. This means that the properties of those wholes are determined by the properties and relations of their proper parts.

3:AM:  One of the things you metaphysicians try and do is carve nature at the joints in various ways.
You suggest that doing this normatively has advantages over other approaches. Can you sketch what it means to carve nature at the joints normatively and why it’s so useful? Does this mean that some sort of morality gets to exist at a fundamental level, which would be a pretty significant discovery wouldn’t it?

KM: Right, the notions of fundamentality and carving at the joints have already been mentioned a couple of times above. As I mentioned too, the theory of fundamentality I favor is that how metaphysically fundamental a thing is just is how real that thing is. But there are other plausible theories out there, and one of them, which I explore in a recent paper, accounts for fundamentality in normative terms. I’ll talk more about this normative theory of fundamentality in a moment but first let me say a bit more about what led me to consider this theory in the first place.

First, I was thinking about a medieval tradition according to which absolutely everything has a kind of goodness that is identical with its being, albeit under a different guise. (The concept of goodness isn’t the same concept as the concept of being, but the underlying phenomenon both refer to is the same.) Let’s call this kind of goodness metaphysical goodness. (If there are modes of being, there I assume that there would also be modes of metaphysical goodness, but let’s set aside this complication for now.) The metaphysical goodness of a thing is proportionate to how much being that thing has. Since I like degrees of being, I started wondering whether and how to make sense of metaphysical goodness.

Second, I was thinking about a family of meta-ethical views on which goodness is to be understood in terms of something normative. One version of this view is often ascribed to Brentano, according to which to be intrinsically good just is to be the correct object of an (intrinsic) act of love. Let me give some examples to make this idea more concrete and down to earth. Consider two different instances in which someone is pleased that something is the case. First, a real life example: when my older daughter was four, we had a conversation about how condensation works. This conversation included a demonstration in which I took a plate that had been in the freezer for a few minutes and placed it above a pot of boiling water. She got a verbal explanation of condensation and a visual demonstration. The next day she demonstrated that she understood the idea by applying it in a new situation: she saw the condensed water on the car in the morning. She then explained to me how condensation works. It blew my mind, and convinced me that it was always worth the effort to try to explain things to her if I can. I also felt an enormous amount of pleasure, the object of which was her understanding of condensation. This is the sort of pleasure—pleasure taken in the innocent knowledge of a child—that Brentano would call correct and that others in this tradition would call a pleasure taken in a fitting or appropriate object. Now for the second example. Suppose instead of taking pleasure in the innocent knowledge of a child, one took pleasure in her underserved suffering. The object of that pleasure would not be fitting, and the pleasure taken in it would be incorrect. Brentano also seemed to understand being intrinsically better than in terms of correctness of preference: for one thing to be intrinsically better than another is for a preference of the first over the second to be correct.

This account of intrinsic value in terms of correctness is an instance of a family of views, each of which understand a kind of value in terms of something normative, like correctness, obligation, reasons, or whatnot.

If you put the first and second thoughts together, you get an interesting result: if being is a kind of metaphysical goodness, then there must be a connection between being and normativity, since types of goodness ultimately are understood in terms of types of normativity. Naturally, I wondered whether we could give an account of fundamentality (and maybe then being itself?) in normative terms. So that’s how and why I started thinking about the view that I discuss in my paper “Normative Accounts of Fundamentality”.

Incidentally, one reason to read widely is that there can be neat connections between what seem to be completely unrelated topics.

Roughly, the basic idea of the paper is that there are prima facie obligations to theorize about certain properties, and that the grade of fundamentality of a property consists in how strong that prime facie obligation is. Throughout the paper, this rough and basic idea is refined, and I’ll spare you the recapitulation of that refinement here!

I don’t think the view implies that there is morality at the fundamental level. For one thing, the view is consistent with emotivism (or other versions of non-cognitivism) about normativity, and these theories preclude the possibility of normativity at any level of reality, let alone the fundamental level. Emotivism is the theory that normative utterances are really just expressions of emotion. On this theory, if I say “murdering your neighbors is morally wrong”, what I am really saying is something that “Boo to murdering your neighbors!” or “Murdering your neighbors—yuck!”. I’m not saying something that can be true or false because I am not asserting a proposition that might match or fail to match reality. I’m just venting. Emotivism is the “commenting on internet forums” theory of normative language. Emotivism is a kind of non-cognitivism—I hesitate to try to define “non-cognitivism” because what we should(!) mean by that term has become very contested, but for our purposes just think of it as a view on which normative utterances don’t have the function of describing reality, but rather serve some other purpose.

The emotivist or non-cognitvist could accept the account I present, and thereby have an emotivist or non-cognivist account of metaphysical fundamentality. It’s interesting to think of whether they could take the further step of identifying being and metaphysical goodness—arguably not, since the product of their joining would be a non-cognivitist view of being itself!

So the view doesn’t entail that there are fundamental normative properties. But if realism about normativity is true, there’s a good case to be made on the basis of this view for the fundamentality of the normative property used to account for fundamentality itself. It is hard to see how fundamentality could play an important role in metaphysics unless fundamentality is itself highly fundamental. Ted Sider has argued that fundamentality must itself be fundamental in order to do the work it is called on to do. I don’t accept this argument, but I agree at least that fundamentality must be highly fundamental. And this means that the properties in terms of which fundamentality is accounted for must themselves be highly fundamental. This does seem like a significant consequence if it is correct.

One reason this view is so intriguing is that so many meta-ethicists struggle to “locate” moral properties in the non-moral world. They don’t just let moral properties and so on be “their own thing”. I think part of the reason is that they buy in to an antecedent commitment that some properties are more fundamental than others—a commitment I share—and they think that moral properties can’t be fundamental. But if fundamentality is to be accounted for normatively—if fundamentality is thereby itself a normative property—then we need to rethink the terms of the game.

3:AM:  Are there parts of reality that are eternal? I guess I’m interested in just what an overview of your ontology looks like – is it a desert landscape like van Inwagen’s – just concrete objects (including supernatural objects) and abstracts (which is where the eternals are I would suppose) – or very different from that?

KM: I believe only in water. (Just kidding.) My fundamental ontology—the list of those things that are fully real/have the maximum amount of being—is a lot like van Inwagen’s. An important difference though is that van Inwagen doesn’t distinguish between fundamental ontology in this sense and a list of whatever has some mode or amount of being, because he views this distinction as unintelligible. He distinguishes only what there is from what there isn’t, so to speak, whereas I see shades and grades of being.

I don’t know how to prove that there are fully real things, but it is a reasonable assumption to make. I don’t think it is a mandatory assumption though. Maybe part of the claim that everything is empty made by the Mahayana school of Buddhism is that nothing is fully real in the sense I intend.

Given that there are fully real things, the task is to identify them. The entities accounted for in the physical and biological sciences seem like plausible candidates. But so to do the entities accounted for in the mathematical ones. (And maybe also the ones accounted for in the moral sciences!) I don’t think that there are great arguments for a strongly physicalistic worldview.

3:AM:  And finally, how do we decide whether a metaphysical system is the right one or not? Empirical evidence doesn’t seem to be decisive, and given that some metaphysicians have impossible worlds and Meinongian objects it seems anything might happen in such a system. Are all metaphysical issues undecidable – and extending this thought to philosophy generally do you agree with the thought that philosophical questions are also generally without a decisive right answer, that every and any philosophical question can be reopened and rethought?

KM: This is a very hard question. I wouldn’t say that all metaphysical questions are undecidable. I know that I exist, and I think I have very good evidence that I am a material object who is made of other material objects. So I have at least one good reason to believe that compositional nihilism is false. Is this a conclusive argument? It probably won’t convince the diehard compositional nihilist! But that doesn’t mean it is unsound, and it doesn’t mean that I don’t have a reason to believe the conclusion on the basis of that argument. But since I also think that being comes in degrees, there is a further question about how real I am that isn’t settled by this argument. Answering this further question is much harder I think.

The moral I draw from this is that we shouldn’t think that all metaphysical beliefs are epistemically on a par. Some of them are more epistemically justified than others. The question of what in general can justify metaphysical beliefs is also a hard question. (I’m not sure that there is a special problem concerning how to justify metaphysical beliefs as opposed to philosophical beliefs more generally.)

One might be tempted to think that there is endless disagreement among metaphysicians. But that endless disagreement takes place against a backdrop of widespread and mostly unacknowledged agreement. I think that this backdrop has at least two sources.

Some of this agreement might be the result of culpably ignoring alternative views that should be taken seriously, or the philosophical milieu interlocutors are in. Academic philosophy is, for better or worse, subject to fads too, and the increasing arms race to publish might make this worse.

We should diligently look for our philosophical blind spots. The history of philosophy, cross-cultural philosophy, and experimental philosophy can each help with doing this. Attending to disciplines outside of philosophy can help. So can thinking carefully about the space of answers to philosophical questions even when (maybe even especially when) one is not tempted by a specific answer.

But I don’t think the totality of the backdrop of widespread and mostly unacknowledged agreement can be explained wholly by the factors just mentioned. Consider a specific metaphysical question, such as the question of when composition occurs. There are many answers to this question defended in the extant literature. But no one defends the answer that some objects compose a whole if and only if they are in New Jersey. No one defends the answer that some objects compose a whole if and only if there are exactly 15 of them. No one defends the answer that some objects compose a whole if and only if Donald Trump believes that they compose a whole. These answers are nonstarters. They are so clearly false that they don’t even need to be explicitly formulated in order to be rejected. We know that these answers are false. There is widespread, tacit knowledge about what sort of answers to metaphysical questions can be rejected out of hand. Thinking clearly about how we have this knowledge would be worthwhile.

That we can rule out some answers to metaphysical questions fairly decisively doesn’t mean that we can rule in exactly one. With respect to questions in fundamental metaphysics, disciplined and humble speculation might be the best we can hope for.

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

KM: This is a surprisingly hard question to answer. I have a seriously bad case of bibliomania. It started in college, got worse in graduate school, and really got out of control when I got a job.

The first two books of contemporary metaphysics I read were Peter van Inwagen’s Material Beings and

David Lewis’s On the Plurality of Worlds. Pretty much right after I declared a philosophy major, I asked the more senior philosophy majors what to read, and these were the books which they said everyone was talking about. So I checked them out from the library and, over the next few weeks, read them both cover to cover. Material Beings is a very accessible book, although some portions of it are going to be tough for someone with not a lot of philosophy under his or her belt. Material Beings occupied my thoughts for months and months after, and I probably bored or freaked out a lot of my friends by incessantly talking about it. Reading Plurality of Worlds was not as initially rewarding. Maybe I understood about two percent of it, and the feeling of being so completely out of my depth was scary, exciting, and often demoralizing. But it is a book that I have continued to return to, and I love teaching it as well. So these two books should be on my list.

I like this remark by Foucault about how it is important to have a small number of authors with whom one thinks, with whom one works, but on whom one doesn’t write. For him, those authors included Heidegger and Nietzsche. For me, Leibniz and Kant would be in that group, except that I do occasionally write on Kant, so maybe including him is a bit of a cheat.

But let’s put Kant’s Critique of Pure Reason on the list.

Now I’m going to really cheat and give you three more book recommendations, all published relatively recently.

First, since the history of philosophy has provided me with a lot of inspiration for my contemporary work, I’ll mention that I loved Robert Pasnau’s Metaphysical Themes 1274-1671. It’s a real page-turner that covers centuries of awesome metaphysics. And Pasnau is an exceptionally clear writer. You can learn a lot of metaphysics from this book.

Second, I find it very inspiring when a philosopher opens up new intellectual territory, and in that vein, I’ll mention Laurie Paul’s Transformative Experience, which is just super cool.

Third, I just finished Bryan van Norden’s Taking Back Philosophy, and while I am not fully convinced by everything he says, the book makes a strong case for making major changes in what is taught and studied in most philosophy departments in “the West”. It has motivated me to do better too.


Richard Marshall is still biding his time.

Buy his new book here or his first book here to keep him biding!

End Times Series: the first 302

First published in 3:AM Magazine: Saturday, April 7th, 2018.