Donald L. M. Baxter

To the question whether Hume believed in mind-independent physical objects (or as he would put it, bodies), the answer is Yes and No. It is Yes when Hume writes “We may well ask, What causes induce us to believe in the existence of body? but ’tis in vain to ask, Whether there be body or not? That is a point, which we must take for granted in all our reasonings.” However the answer is No after inquiring into the causes of that belief, when he writes, “What then can we look for from this confusion of groundless and extraordinary opinions but error and falsehood? And how can we justify to ourselves any belief we repose in them?” My view is that Hume’s Yes and No answers can best be accounted for by seeing him as a Pyrrhonian skeptic in the tradition of Sextus Empiricus. This interpretation, while finding considerable truth in the various positions canvassed, is closest to those of Norton, Fogelin, and especially Popkin.

Hume’s empiricist reason for rejecting “school metaphysics” makes it natural to assume that Hume rejects all metaphysics. A.J. Ayer certainly reads Hume this way. The natural assumption is wrong, however. Hume only rejects the aprioricity of metaphysics, and not the science itself. I will argue that his empirical science of human nature supports three basic metaphysical principles. (1) The Contradiction Principle: The clearly conceivable implies no contradiction. (2) The Conceivability Principle: The clearly conceivable is possible. (3) The Conceptual Separability Principle: Things are different if and only if distinguishable, and distinguishable if and only if separable in conception. On these principles the rest of Hume’s metaphysics is based, including his atomistic ontology and his denial of necessary connection. And so, for Hume metaphysics is an empirical science.

Hume’s critique of traditional abstraction entails a result that undercuts his account of the idea of identity. To save his account of identity, Hume would have to accept abstraction as well. What links these two discussions is (1) Hume’s widely shared assumption that traditional abstraction is separating in the mind what are inseparable in reality, (2) his principle that what are different are mentally separable, and (3) his principle that we cannot conceive of the impossible. Given these, it will turn out that abstraction is mentally separating something from itself, which will entail that abstraction is conceiving of something as distinct from itself. But it is impossible for something to be distinct from itself, and so it is inconceivable. Therefore abstraction is impossible. Yet consider Hume’s account of the idea of identity. On that account, to conceive of an identity is to conceive of something as one single thing viewed one way and as two distinct things viewed another. How we can take these opposing views of the same thing is a problem that I’ve termed Hume’s Difficulty concerning Identity. It will turn out that we can take the opposing views only if we can conceive of the single thing viewed one way as somehow distinct from itself when viewed the other way. That is, we must be able to conceive of something as distinct from itself. However, if we cannot conceive of something as distinct from itself when abstracting, then we cannot do it when conceiving of an identity. So traditional abstraction and Hume’s account of the idea of identity stand or fall together.

Stefanie Rocknak has written an ambitious and challenging book1 in which she argues for a new interpretation of Hume's account of how we come to believe in external objects, and what it is we believe in. I am hampered by the fact that she and I seem to agree on so little. Thus, my criticisms run the danger of simply not seeing what she is up to.A preliminary terminological point: where Rocknak uses the word "object," I will often use the word "body," since I think Hume sometimes uses "object" in a more general sense that includes perceptions.If I understand correctly, Rocknak's Hume argues that we come to believe in bodies through a special kind of causal reasoning. We reason what must exist in order to cause...

There is more than one way to say that composition is identity. Yi has distinguished the Weak Composition thesis from the Strong Composition thesis and attributed the former to David Lewis while noting that Lewis associates something like the latter with me. Weak Composition is the thesis that the relation between the parts collectively and their whole is closely analogous to identity. Strong Composition is the thesis that the relation between the parts collectively and their whole is identity. Yi is right that Strong Composition does not fully reflect my view. We must recognize further what, borrowing Lewis's characterization, could be called the "stronger, and stranger" Composition thesis, or what I'll call the Stronger Composition thesis for short. On Strong Composition, it is only collectively that the parts are identical with the whole. On Stronger Composition, they are individually identical with it as well. I will explain and motivate this thesis.

Is a social complex identical to many united people or is it a group entity in addition to the people? For specificity, I will assume that a social complex is a plural subject in Margaret Gilbert’s sense. By appeal to my theory of Aspects, according to which there can be qualitative difference without numerical difference, I give an answer that is a middle way between metaphysical individualism and metaphysical holism. This answer will enable answers to two additional metaphysical questions: (i) how can two social complexes have all the same members and (ii) how can there be a social complex of social complexes?

Confucius gave counsel that is notoriously hard to follow: "What you do not wish for yourself, do not impose on others" (Huang 1997: 15.24). People tend to be concerned with themselves and to be indifferent to most others. We are distinct from others so our self-concern does not include them, or so it seems. Were we to realize this distinctness is merely apparent--that our true self includes others--Confucius's counsel would be easier to follow. Concern for our true self would extend concern beyond the narrow selves we appear to be. The neo-Confucians held just such a view. They espoused an identity with the universe and everything in it, arguing that this identity explains a natural concern for everyone and everything, not just for our narrow selves. However, many things in the universe differ from each other, that is, some have qualities others lack. If they are all one and the same thing then that one thing differs from itself. I will suggest that the objection can be answered with some metaphysical innovation. I will address the objection by sketching a theory--call it the theory of aspects--that explains how numerically identical things can differ qualitatively.

ABSTRACT An apparent objection against my theory of instantiation as partial identity is that identity is necessary, yet instantiation is often contingent. To rebut the objection, I show how it can make sense that identity is contingent. I begin by showing how it can make sense that identity is temporary. I rely heavily on Andre Gallois’s formal theory of occasional identity, but argue that there is a gap in his explanation of how his formalisms make sense that needs to be filled by appeal to my theory of Aspects.

The ancient theory of substance and accident is supposed to make sense of complex unities in a way that respects both their unity and their complexity. On Hume’s view such complex unities are only fictitiously unities. This result follows from his thoroughgoing critique of the theory of substance. I will characterize the theory Hume is critiquing as it is presented in Locke, presupposing what Bennett calls the “Leibnizian interpretation.” Locke uses the word ‘substance’ in two senses. Call substance in the first sense “individual substance” and in the second sense “pure substance.” I will discuss the seven main parts of Hume’s view: (I) that we have no idea of pure substance, (II) that there is no complex individual substance, except in a loose sense, (III) that the fiction of complex individual substance arises in a way parallel to that of the fiction of identity through time, and (IV) results in the fiction of pure substance, (V) that simple qualities and perceptions satisfy the definition of individual substance, (VI) that there is no such thing as inherence, and (VII) that there is no such thing as pure substance. Hume’s views on substance are often mentioned without being discussed in detail. Kemp Smith, Stroud, and Garrett, for example, mostly summarize various claims of Hume in the course of expounding on his theory of the idea of personal identity. In contrast, I will attempt to present a systematic treatment of Hume on substance as a refutation of Locke.

Oliver's and Rodriguez-Pereyra's important interpretation of the problem of universals as one concerning truthmakers neglects something crucial: that there is a numerical identity between numerically distinct particulars. The problem of universals is rather how to resolve the apparent contradiction that the same things are both numerically distinct and numerically identical. Baxter's account of instantiation as partial identity resolves the apparent contradiction. A seeming objection to this account is that it appears to make instantiation symmetric, since partial identity is symmetric. Armstrong's standard reply is that the difference between a particular and a universal is what makes instantiation asymmetric. Brown suggests, though, that the instantiation of a universal by a universal is sometimes symmetric. However, the examples on which he relies are not universals.

ABSTRACT According to David Lewis, alteration is "qualitative difference between temporal parts of something." It follows that moments, since they are simple and lack temporal parts, cannot alter from future to present to past. Here then is another way to put McTaggart's paradox about change in tense. I will appeal to my theory of Aspects to rebut the thought behind this rendition of McTaggart. On my theory, it is possible that qualitatively differing things be numerically identical. I call these differing, numerically identical things "aspects." I will argue that alteration can be a qualitative difference between temporal aspects of something that lacks temporal parts. So a moment can alter in tense. By rejecting Lewis's assumption my theory can solve this version of McTaggart's paradox.

In this volume--the first, focused study of Hume on time and identity--Baxter focuses on Hume’s treatment of the concept of numerical identity, which is central to Hume's famous discussions of the external world and personal identity. Hume raises a long unappreciated, and still unresolved, difficulty with the concept of identity: how to represent something as "a medium betwixt unity and number." Superficial resemblance to Frege’s famous puzzle has kept the difficulty in the shadows. Hume’s way of addressing it makes sense only in the context of his unorthodox theory of time. Baxter shows the defensibility of that theory against past dismissive interpretations, especially of Hume’s stance on infinite divisibility. Later the author shows how the difficulty underlies Hume’s later worries about his theory of personal identity, in a new reading motivated by Hume’s important appeals to consciousness. Baxter casts Hume throughout as an acute metaphysician, and reconciles this side of Hume with his overarching Pyrrhonian skepticism.

We ordinarily make statements of the form "They are the same thing," if there has been reason to distinguish what we now judge identical. But such statements seem not to make sense. "They" indicates that there are more than one thing, whereas "same" indicates that there is only one thing. How can many be one? Hume's obscure Principle of Identity passage in the Treatise addresses this problem . Call it the Number Problem for Identity. Clarifying Hume's account reveals that, despite its richness and suggestiveness, it is ultimately unsatisfactory. In discussing identity, he tries to solve an apparent number problem at the level of objects of ideas, only to leave himself with a real one at the level of ideas. He tries to solve the number problem for identity by appeal to one continuous whole and its many continuous parts, only to leave himself with a different number problem for parts/whole. He tries to make discontinuity necessary for numerical difference, but neither are. ;I develop his account into a satisfactory one, influenced by British Idealist discussions of Identity in Difference and by Hume's discussion of Distinctions of Reason. In response to I propose the concept of "Difference in Respect" based on the dictum that it is possible for an object in different respects to have and lack a property. In the problematic identity attribution I began with, "they" refers to the object in certain various respects, while "same" indicates that it is one object. This is a solution to the number problem for identity. In response to I solve the number problem for parts/whole by suggesting that parts and whole exist and count in different contexts. Which context is the better one to count in varies with purpose and interest. In response to I suggest that Hume's supposedly necessary conditions for numerical distinctness become criteria for judging between contexts, e.g. for many purposes it is better to count a single continuous whole rather than its many continuous parts. I end by applying my newly developed theory to other problems. I show how the same thing can have or lack a property depending on contexts determined by interest or purpose

Armstrong has loose identity be an equivalence relation, yet in cases of something becoming something else, loose identity is not transitive. My alternate account has an attribution of loose identity be really two: a true attribution of an underlying relation (perhaps not transitive) and a false attribution--a Humean feigning-of strict identity. The feigning may become less appropriate as the underlying relation grows more distant. What makes it appropriate initially is that the underlying relation supports a predictable change in some collective. The importance of the predictably changing collective is signaled by regarding it as a single thing.

Hume’s account of virtue (and beauty) entails that distinct things--a quality in the contemplated and a perception in the contemplator--are the same thing--a given virtue. I show this inconsistency is consistent with his intent. A virtue is a composite of quality and perception, and for Hume a composite is distinct things--the parts--falsely supposed to be a single thing. False or unsubstantiated supposition is for Hume the basis of most of our beliefs. I end with an argument that for Hume secondary qualities are composites, not powers.

Understanding Hume’s theory of space and time requires suspending our own. When theorizing, we think of space as one huge array of locations, which external objects might or might not occupy. Time adds another dimension to this vast array. For Hume, in contrast, space is extension in general, where being extended is having parts arranged one right next to the other like the pearls on a necklace. Time is duration in general, where having duration is having parts occurring one aft er another like the notes of a song. Hume’s diff erent view stems from his empiricism, his reliance on experience and observation as the foundation of our concepts. Nothing in our experience suggests a single vast array of locations. Rather, we simply notice that bodies are similar insofar as they have lengths that can be compared. Likewise, nothing in our experience suggests a single dimension of time. Rather, we simply notice that diff erent successions are similar insofar as they have durations that can be compared. Th eorizing that these observations show there to be a single multidimensional array goes well beyond the evidence for Hume. As a skeptic, he fi nds himself unable to assent to theories that stray too far beyond the deliverances of the senses. For Hume, the ideas of space and time are each a general idea of simple—partless— objects arrayed in a certain manner. He argues that the structures of the ideas of space and time refl ect the structures of space and time. Th erefore, space and time are not infi - nitely divisible, and they are ways simple objects are arrayed. Consequently, there is no such thing as empty space nor time without change.

I argue that an individual has aspects numerically identical with it and each other that nonetheless qualitatively differ from it and each other. This discernibility of identicals does not violate Leibniz's Law, however, which concerns only individuals and is silent about their aspects. They are not in its domain of quantification. To argue that there are aspects I will appeal to the internal conflicts of conscious beings. I do not mean to imply that aspects are confined to such cases, but the best way to start is to recognize them experientially. We can feel the conflicts within ourselves. In doing so we can feel some of our aspects. I will try to enhance our understanding of the concept of aspect by listing and formalizing some principles for its use. After that I will argue that all sorts of individual things have aspects, not just people who are conflicted.

One of the advantages of my account in the essay “Instantiation as Partial Identity” was capturing the contingency of instantiation—something David Armstrong gave up in his experiment with a similar view. What made the contingency possible for me was my own non-standard account of identity, complete with the apparatus of counts and aspects. The need remains to lift some obscurity from the account in order to display its virtues to greater advantage. To that end, I propose to respond to those who have grappled with it in print. There are various criticisms by commentators: that it is rendered absurd by the transitivity of identity, that it makes instantiation necessary instead of contingent, that it is unclear what counts are, that aspects are simply tropes, that my view does not capture multiple location, that I make an unclear reference to a theory of composition as identity, that the account suffers from problems with polyadicity, and that it is not a realist account of universals after all. I give responses to these objections

In the Treatise Hume argues that the self is really many related perceptions, which we represent to ourselves as being one and the same thing. In the Appendix he finds this account inconsistent. Why? The problem arises from Hume's theory that representation requires resemblance. Only a many can represent a many recognized as such, and only a one can represent something as one. So for the many distinct perceptions (recognized as such) to be represented as one and the same, the many distinct ideas that reflect them must be one and the same. But the distinct cannot be identical

Berkeley and Hume object to Locke's account of abstraction. Abstraction is separating in the mind what cannot be separated in reality. Their objection is that if a is inseparable in reality from b, then the idea of a is inseparable from the idea of b. The former inseparability is the reason for the latter. In most interpretations, however, commentators leave the former unexplained in explaining the latter. This article assumes that Berkeley and Hume present a unified front against Locke. Hume supplements Berkeley's argument just where there are gaps. In particular, Hume makes explicit something Berkeley leaves implicit: The argument against Locke depends on the principle that things are inseparable if and only if they are identical. Abstraction is thinking of one of an inseparable pair while not thinking of the other. But doing so entails thinking of something while not thinking of it. This is the fundamental objection

Zu Arnauld und im Discours de métaphysique sagt Leibniz, daß alle Wahrheiten begrifflich (prädikativ) und manche gleichwohl kontingent sind. Ich untersuche das Problem im Hinblick auf mögliche Wesen, die ich als möglich auch betrachte und versuche nachzuweisen, daß die Position keinen Widerspruch enthält, weil Leibniz zwei Arten begrifflichen Enthaltenseins unterscheidet -logisch und kausal: Die erste ist notwendig, die zweite jedoch kontingent und nur hypothetisch notwendig, notwendig also lediglich unter der Voraussetzung des vorgegebenen freien Willens Gottes. Es gibt insofern auch zwei Arten vollständiger Begriffe - logisch und kausal vollständig: Kausal vollständige Begriffe sind logisch unvollständig und enthalten gleichwohl logisch (hypothetisches Enthaltensein) Gründe für die fehlenden Prädikate. Diese Begriffe sind Begriffe von zeitlichen Zuständen, die in einem Individuum durch Gottes wohlbegründete freie Ratschlüsse verbunden sind - solche Zustände lassen sich zwar separieren, jedoch nicht neu zusammenfügen. Diese Interpretation ist vereinbar mit dem inesse-Prinzip, mit Gottes Wissen um zukünftige zufällige Ereignisse sowie Leibniz' Ablehnung des Konzepts weltübergreifender Identität und kann durch die Einbeziehung des Kontingenz-Kriteriums einer unendlichen Analyse sowie einer Theorie der Gegenstücke zur Erklärung des Kontrafaktischen problemlos erweitert werden

In lieu of an abstract, here is a brief excerpt of the content:323 A DEFENSE OF HUME ON IDENTITY THROUGH TIME A durable complaint against Hume is that he blatantly begs the question in his Treatise account of our acquisition of the idea of identity through time. Green and Grose made the accusation in 1878; one hundred years later Stroud echoed the same accusation, its force and liveliness seemingly undiminished. I suggest that this accusation is based on a tempting but unwarranted assumption about Humean perceptions. It is unwarranted despite Price's arguments for it. When the assumption is corrected Hume is exonerated. This defense prepares the way for a new understanding of Hume's subtle and important theory of identity. Clarifying his theory is the goal; defending him is the means. I will merely summarize the Green and Grose interpretation, saving detailed examination for Stroud's. Stroud's version seems more tightly argued for and spares one the problems of interpreting commentators from a different philosophical tradition. Green and Grose run together Hume's account of the acquisition of the idea of identity, with his account of the application of that idea to a succession of related perceptions. This fully intended confusion can be explained by their understanding of Hume's claim that "impressions are... perishing existences." (T 194) They assume that for Hume no perception (impression or idea) is a unity if it exists at an interval of time composed of a succession of times. Rather the perception is really a succession of perceptions. Their assumed interpretation makes Hume seem to them to beg the question as follows: To acquire the idea of identity through time is to acquire the idea of something 324 which is unitary and exists at a succession of times. But any impression which might give this idea is really a succession of related impressions. A succession of related impressions can convey the idea of identity only by being confused with a unitary impression which exists at a succession of times. Such confusion would be possible only for someone who already had the idea of something unitary which exists at a succession of times. So the idea of 2 identity can be acquired only if it is already had. In other words, Hume claims that an impression had for a relatively long time gives the idea of identity. Any long impression is really a succession. A succession can give the idea of identity only if it is confused with something that has identity. That confusion requires the idea of identity already. So Hume begs the question. There are gaps in this argument. But its influence has been important and its structure is clear: If we assume that perceptions are exceedingly brief, then Hume begs the question. (I will call the assumption 'the prosecution's assumption.') Stroud's argument that Hume begs the question has the same structure, or so I will argue. The role of the prosecution's assumption is not immediately apparent, but it is a crucial role. Stroud's accusation is as follows: By Hume's definition, identity is the invariableness and uninterruptedness of an object through a supposed change in time. (T 201) However, on Hume's account of acquiring the concept of identity, the acquisition involves imagining a change in time without any variation or interruption in an object. But that just ¿s, by definition, to imagine the object's identity through the change in time. So acquiring the concept of identity requires an act of imagination involving the 325 concept of identity. So, Stroud concludes, on Hume's account the concept of identity can be acquired only 3 if it is had. Hume's account begs the question. I think it is usually unwise to attribute blunders to subtle thinkers. I will not do it to Hume, nor to his commentators. I will argue that what Hume presupposes in explaining identity through time is a little different from identity through time. Thus the presupposition is not question-begging. More specifically, I will argue that Hume distinguishes three technical concepts, three ways to take up time: (i) merely occupying an interval of time, (ii) having duration for an interval of time, and (iii) being identical through an interval of time. Concept (iii...

I argue via examples that there are cases in which things that are not two distinct things qualitatively differ without contradiction. In other words, there are cases in which something differs from itself. Standard responses to such cases are to divide the thing into distinct parts, or to conceive of the thing under different descriptions, or to appeal to different times, or to deny that the property had is the property lacked. I show these responses to be unsatisfactory. I then gather and systematize available ways of talking about such cases with phrases like ‘insofar as’, ‘qua’, ‘to the extent that’, ‘in some respect’, etc., while paying special attention to the scope of ‘not’ when used with these phrases. This allows me to show how we can speak of self-differing without contradiction.