| 5015 | A universal is a single idea applied to individual things that are similar to one another [Descartes] |
| Full Idea: Universals arise solely from the fact that we avail ourselves of one idea in order to think of all individual things that have a certain similitude. When we understand under the same name all the objects represented by this idea, that name is universal. | |
| From: René Descartes (Principles of Philosophy [1646], I.59) | |
| A reaction: Judging by the boldness of the pronouncement, it looks as if Descartes hasn't recognised the complexity of the problem. How do we spot a 'similarity', especially an abstraction like 'tool' or 'useful'? This sounds like Descartes trying to avoid Platonism. |
| 6714 | Universals do not have any intrinsic properties, but only relations to particulars [Berkeley] |
| Full Idea: Universality, so far as I can comprehend it, does not consist in the absolute, positive nature or conception of anything, but in the relation it bears to the particulars signified or represented by it. | |
| From: George Berkeley (The Principles of Human Knowledge [1710], Intro §15) | |
| A reaction: I always think it is a basic principle in philosophy that some sort of essence must precede relations (and functions). What is it about universals that enables them to have a relation to particulars? |
| 11098 | Momentary impressions are wrongly identified with one another on the basis of resemblance [Hume, by Quine] |
| Full Idea: Momentary impressions, according to Hume, are wrongly identified with one another on the basis of resemblance. | |
| From: report of David Hume (Treatise of Human Nature [1739]) by Willard Quine - Identity, Ostension, and Hypostasis 3 | |
| A reaction: I don't have a Hume quotation for this yet, but Quine is plausibly claiming Hume as a resemblance nominalist, equipped with an error theory about universals. |
| 7954 | If we see a resemblance among objects, we apply the same name to them, despite their differences [Hume] |
| Full Idea: When we have found a resemblance among several objects, that often occur to us, we apply the same name to all of them, whatever differences we may observe in the degrees of their quantity and quality, and whatever other differences may appear among them. | |
| From: David Hume (Treatise of Human Nature [1739], I.I.7) | |
| A reaction: This must to some extent by right, whatever objections can be found. Russell's objection (Idea 4441) wouldn't alter the truth of Hume's observation, thought Hume is attacking universals and Russell defending them. |
| 23649 | No one thinks two sheets possess a single whiteness, but all agree they are both white [Reid] |
| Full Idea: If we say that the whiteness of this sheet is the whiteness of another sheet, every man perceives this to be absurd; but when he says both sheets are white, this is true and perfectly understood. | |
| From: Thomas Reid (Essays on Intellectual Powers 5: Abstraction [1785], 3) | |
| A reaction: Well said. Only a philosopher could think the whiteness of one sheet is exactly the same entity as the whiteness of a different sheet. We seem to have brilliantly and correctly labelled them both as white, and then thought that one word implies one thing. |
| 4441 | 'Resemblance Nominalism' won't work, because the theory treats resemblance itself as a universal [Russell] |
| Full Idea: To be a universal, a resemblance must hold between many pairs of white things. We can't say there is a different resemblance between each pair, since the resemblances must resemble each other, so we are forced to admit that resemblance is a universal. | |
| From: Bertrand Russell (Problems of Philosophy [1912], Ch. 9) | |
| A reaction: Apparently this objection is much discussed and controversial. It looks like a threat to any theory of universals (involving 'sets', or 'concepts', or 'predicates'). We seem to need 'basic' and 'derivative' universals. Cf Idea 7956. |
| 7956 | If all and only red things were round things, we would need to specify the 'respect' of the resemblance [Goodman, by Macdonald,C] |
| Full Idea: According to Goodman's 'companionship difficulty', resemblance nominalism has a problem if, say, all and only the red things were the round things, because we cannot distinguish the two different respects in which the things resemble one another. | |
| From: report of Nelson Goodman (The Structure of Appearance [1951]) by Cynthia Macdonald - Varieties of Things Ch.6 | |
| A reaction: Goodman opts for extreme linguististic nominalism in response to this (Idea 7952), whereas Russell opts for a sort of Platonism (4441). The current idea gives Russell a further problem, of needing a universal of the respect of the resemblance. |
| 7957 | Without respects of resemblance, we would collect blue book, blue pen, red pen, red clock together [Goodman, by Macdonald,C] |
| Full Idea: Goodman's 'imperfect community' problem for Resemblance Nominalism says that without mention of respects in which things resemble, we end up with a heterogeneous collection with nothing wholly in common (blue book, blue pen, red pen, red clock). | |
| From: report of Nelson Goodman (The Structure of Appearance [1951]) by Cynthia Macdonald - Varieties of Things Ch.6 | |
| A reaction: This suggests Wittgenstein's 'family' resemblance as a way out (Idea 4141), but a blue book and a red clock seem totally unrelated. Nice objection! At this point we start to think that the tropes resemble, rather than the objects. |
| 8532 | Resemblances must be in certain 'respects', and they seem awfully like properties [Armstrong] |
| Full Idea: If a resembles b, in general, they resemble in certain respects, and fail to resemble in other respects. But respects are uncomfortably close to properties, which the Resemblance theory proposes to do without. | |
| From: David M. Armstrong (Properties [1992], §1) | |
| A reaction: This is a good objection. I think it is plausible to build a metaphysics around the idea of respects, and drop properties. Shall we just talk of 'respects' for categorising, and 'powers' for causation and explanation? Respects only exist in comparisons. |
| 4439 | 'Resemblance Nominalism' says properties are resemblances between classes of particulars [Armstrong] |
| Full Idea: Resemblance Nominalists say that to have a property is to be a member of a class which is part of a network of resemblance relations with other classes of particulars. ..'Resemblance' is taken to be a primitive notion, though one that admits of degrees. | |
| From: David M. Armstrong (Universals [1995], p.503) | |
| A reaction: Intuition suggests that this proposal has good prospects, as properties are neither identical, nor just particulars, but have a lot in common, which 'resemblance' captures. Hume saw resemblance as a 'primitive' process. |
| 4440 | 'Resemblance Nominalism' finds that in practice the construction of resemblance classes is hard [Armstrong] |
| Full Idea: It is difficult for Resemblance Nominalists to construct their interconnected classes in practice. | |
| From: David M. Armstrong (Universals [1995], p.503) | |
| A reaction: Given the complexity of the world this is hardly surprising, but it doesn't seem insuperable for the theory. It is hard to decide whether an object is white, or hot, whatever your theory of universals. |
| 8500 | Colour resemblance isn't just resemblance between things; 'colour' must be mentioned [Jackson] |
| Full Idea: Some red things resemble some blue things more than some pink things because of factors other than colour. Nominalists must offer 'anything red colour-resembles anything pink', but that may contain a universal in disguise. | |
| From: Frank Jackson (Statements about Universals [1977], p.90) | |
| A reaction: Hume and Quine are probably right that we spot resemblances instantly, and only articulate the respect of the resemblance at a later stage. |
| 8521 | Nominalism has the problem that without humans nothing would resemble anything else [Campbell,K] |
| Full Idea: The objection to nominalism is its consequence that if there were no human race (or other living things), nothing would be like anything else. | |
| From: Keith Campbell (The Metaphysic of Abstract Particulars [1981], §6) | |
| A reaction: Anti-realists will be unflustered by this difficulty. Personally it strikes me as obvious that some aspects of resemblance are part of reality which we did not contribute. This I take to be a contingent fact, founded on the existence of natural kinds. |
| 7009 | Similarity among modes will explain everthing universals were for [Heil] |
| Full Idea: My contention is that similarity among modes can do the job universals are conventionally postulated to do. | |
| From: John Heil (From an Ontological Point of View [2003], Intro) | |
| A reaction: See Idea 4441 for Russell's nice objection to this view. The very process by which we observes similarities (as assess their degrees) needs to be explained by any adequate theory of properties or universals. |
| 7041 | Similar objects have similar properties; properties are directly similar [Heil] |
| Full Idea: Objects are similar by virtue of possessing similar properties; properties, in contrast, are not similar in virtue of anything. | |
| From: John Heil (From an Ontological Point of View [2003], 14.2) | |
| A reaction: I am not sure if I can understand the concept of similarity if there is no answer to the question 'In what respect?' I suppose David Hume is happy to take resemblance as given and basic, but it could be defined as 'sharing identical properties'. |
| 7720 | Two things can only resemble one another in some respect, and that may reintroduce a universal [Lowe] |
| Full Idea: A problem for resemblance nominalism is that in saying that two particulars 'resemble' one another, it is necessary to specify in what respect they do so (e.g. colour, shape, size), and this threatens to reintroduce what appears to be talk of universals. | |
| From: E.J. Lowe (Locke on Human Understanding [1995], Ch.7) | |
| A reaction: We see resemblance between faces instantly, long before we can specify the 'respects' of the resemblance. This supports the Humean hard-wired view of resemblance, rather than some appeal to Platonic universals. |
| 9447 | If properties are clusters of powers, this can explain why properties resemble in degrees [Mumford] |
| Full Idea: If a cluster of ten powers exhausts property F, and property G differs in respect of just one power, this might explain why properties can resemble other properties and in different degrees. | |
| From: Stephen Mumford (Laws in Nature [2004], 10.6) | |
| A reaction: I love this. The most intractable problem about properties and universals is that of abstract reference - pink resembles red more than pink resembles green. If colours are clusters of powers, red and pink share nine out of ten of them. |
| 4458 | Unlike Class Nominalism, Resemblance Nominalism can distinguish natural from unnatural classes [Moreland] |
| Full Idea: Resemblance Nominalism is clearly superior to Class Nominalism, since the former offers a clear ground for distinguishing between natural and unnatural classes. | |
| From: J.P. Moreland (Universals [2001], Ch.2) | |
| A reaction: Important. It seems evident to me that there are natural classes, and the only ground for this claim would be either the resemblance or the identity of properties. |
| 9472 | Resemblance itself needs explanation, presumably in terms of something held in common [Bird] |
| Full Idea: The realist view of resemblance nominalism is that it is resemblance that needs explaining. When there is resemblance it is natural to want to explain it, in terms of something held in common. Explanations end somewhere, but not with resemblance. | |
| From: Alexander Bird (Nature's Metaphysics [2007], 2.1.2) | |
| A reaction: I smell a regress. If a knife and a razor resemble because they share sharpness, you have to see that the sharp phenomenon falls within the category of 'sharpness' before you can make the connection, which is spotting its similarity. |
| 7955 | Resemblance Nominalism cannot explain either new resemblances, or absence of resemblances [Macdonald,C] |
| Full Idea: Resemblance Nominalism cannot explain the fact that we know when and in what way new objects resemble old ones, and that we know when and in what ways new objects do not resemble old ones. | |
| From: Cynthia Macdonald (Varieties of Things [2005], Ch.6) | |
| A reaction: It is not clear what sort of theory would be needed to 'explain' such a thing. Unless there is an explanation of resemblance waiting in the wings (beyond asserting that resemblance is a universal), then this is not a strong objection. |
| 18435 | Resemblance Nominalists say that resemblance explains properties (not the other way round) [Rodriquez-Pereyra] |
| Full Idea: Resemblance Nominalists cannot explain the resemblance between particulars in terms of their properties, because they explain particulars' properties in terms of their resemblances. | |
| From: Gonzalo Rodriguez-Pereyra (Resemblance Nominalism and Russell's Regress [2001], p.397), quoted by Douglas Edwards - Properties 5.5.1 | |
| A reaction: While resemblance does seem to be a primitive fact of experience, and it points us towards the properties, to say that resemblance explains properties is obviously (as so often...) getting things the wrong way round. Properties ARE resemblances?? |
| 18436 | Entities are truthmakers for their resemblances, so no extra entities or 'resemblances' are needed [Rodriquez-Pereyra] |
| Full Idea: A and B are the sole truthmakers for 'A and B resemble each other'. There is no need to postulate extra entities - the resembling entities suffice to account for them. There is no regress of resemblances, ...since there are no resemblances at all. | |
| From: Gonzalo Rodriguez-Pereyra (Resemblance Nominalism: a solution to universals [2002], p.115), quoted by Douglas Edwards - Properties 5.5.2 | |
| A reaction: This seems to flatly reject the ordinary conversational move of asking in what 'respect' the two things resemble, which may be a genuine puzzle which gets an illuminating answer. We can't fully explain resemblance, but we can do better than this! |
| 18437 | Resemblance nominalism requires a second entity to explain 'the rose is crimson' [Edwards] |
| Full Idea: For resemblance nominalism the sentence 'the rose is crimson' commits us to at least one other entity that the rose resembles in order for it to be crimson. | |
| From: Douglas Edwards (Properties [2014], 5.5.2) | |
| A reaction: If the theory really needs this, then it has just sunk without trace. It can't suddenly cease to be crimson when the last resembling entity disappears. |