structure for 'Reason'    |     alphabetical list of themes    |     unexpand these ideas

2. Reason / D. Definition / 4. Real Definition

[give the true nature of something, not just a description]

25 ideas
A primary element has only a name, and no logos, but complexes have an account, by weaving the names [Plato]
     Full Idea: A primary element cannot be expressed in an account; it can only be named, for a name is all that it has. But with the things composed of these ...just as the elements are woven together, so the names can woven to become an account.
     From: Plato (Theaetetus [c.368 BCE], 202b01-3)
     A reaction: This is the beginning of what I see as Aristotle's metaphysics, as derived from his epistemology, that is, ontology is what explains, and what we can give an account [logos] of. Aristotle treats this under 'definitions'.
A definition must be of something primary [Aristotle]
     Full Idea: We can assign a term to any 'account' whatever, so that the Iliad would be a definition! No, a definition must be of something primary.
     From: Aristotle (Metaphysics [c.324 BCE], 1030a08)
     A reaction: At first hearing you may doubt this claim, but Aristotle's example clinches it beautifully. Are you really going to say that The Iliad is the definition of 'Iliad'?
Only substance [ousias] admits of definition [Aristotle]
     Full Idea: Only substance admits of definition.
     From: Aristotle (Metaphysics [c.324 BCE], 1030b34)
Definitions need the complex features of form, and don't need to mention the category [Aristotle, by Wedin]
     Full Idea: Met. Z.10-11 is committed to the complexity of form and suggests that the complexity is expressed in definitions that articulate, in the case of 'man', the salient faculties and functions, and none of these need mention 'man'.
     From: report of Aristotle (Metaphysics [c.324 BCE], 1034b20-1037b) by Michael V. Wedin - Aristotle's Theory of Substance X.5
     A reaction: This is a very strong statement of the view that identifying genus and species are not at all what Aristotle wants in his final account of essence. The features mentioned here would, though, clearly count as 'differentiae'.
Sometimes parts must be mentioned in definitions of essence, and sometimes not [Aristotle]
     Full Idea: In some instances the account of the what-it-was-to-be-that-thing contains the parts of the thing defined and in others not.
     From: Aristotle (Metaphysics [c.324 BCE], 1037a22)
     A reaction: This is helpful in understanding what an essence is supposed to be. I take it to be sometimes a structure, with parts, and sometimes more like a guiding principle, more abstract in character.
Definitions are of what something is, and that is universal [Aristotle]
     Full Idea: Definitions are thought to be of what something is, and what something is is in every case universal and positive.
     From: Aristotle (Posterior Analytics [c.327 BCE], 90b05)
     A reaction: This is exhibit A for those who think that Aristotelian essences concern the genus, rather than the particular. I suspect that this idea is best expressed as 'all we can say by way of definition of a particular thing involves the use of universals'.
Definition by division needs predicates, which are well ordered and thorough [Aristotle]
     Full Idea: To establish a definition through division, you must aim for three things: you must take what is predicated in what the thing is; you must order these items as first or second; and you must ensure that these are all there are.
     From: Aristotle (Posterior Analytics [c.327 BCE], 97a23)
     A reaction: This gives an indication of the thoroughness that Aristotle expects from a definition. They aren't like dictionary definitions of words. He expects definitions to often be very lengthy (see Idea 12292).
You can define objects by progressively identifying what is the same and what is different [Aristotle]
     Full Idea: Find what is in common among items similar and undifferentiated, then do the same for items of the same kind as the first group but a different form, and so on, till you come to a single account: this will be the definition of the object.
     From: Aristotle (Posterior Analytics [c.327 BCE], 97b07-14)
     A reaction: [His example is distinguishing 'magnanimity' from 'indifference to fortune' among people] Presumably this process works for the formation of new concepts (e.g. in biology), as well as for the definition of familiars in terms of other familiars.
An Aristotelian definition is causal [Aristotle, by Witt]
     Full Idea: An Aristotelian definition is causal.
     From: report of Aristotle (Posterior Analytics [c.327 BCE], Bk II.2) by Charlotte Witt - Substance and Essence in Aristotle 1.5
     A reaction: [She refers us to Posterior Analytics II.2] This is important if we are tempted to follow a modern line of saying that we want Aristotelian essences, and that these are definitions. We ain't thinking of dictionaries.
Aristotelian definitions aim to give the essential properties of the thing defined [Aristotle]
     Full Idea: A real definition, according to the Aristotelian tradition, gives the essence of the kind of thing defined. Man is defined as a rational animal, and thus rationality and animality are of the essence of each of us.
     From: Aristotle (works [c.330 BCE]), quoted by Willard Quine - Vagaries of Definition p.51
     A reaction: Compare Idea 4385. Personally I prefer the Aristotelian approach, but we may have to say 'We cannot identify the essence of x, and so x cannot be defined'. Compare 'his mood was hard to define' with 'his mood was hostile'.
Essential definitions show the differences that discriminate things, and make them what they are [Boyle]
     Full Idea: Essential definitions are such as are taken from the essential differences of things, which constitute them in such a sort of natural bodies, and discriminate them from all those of any other sort.
     From: Robert Boyle (The Origin of Forms and Qualities [1666], p.41?), quoted by Peter Alexander - Ideas, Qualities and Corpuscles
     A reaction: I don't think this goes as far as the aim Aristotle had in definitions, which was more than merely to 'discriminate' each thing. A full definition explains the thing as well.
If our ideas of a thing are imperfect, the thing can have several unconnected definitions [Leibniz]
     Full Idea: The result of having an imperfect idea of something is that the same subject admits of several mutually independent definitions: we shall sometimes be unable to derive one from another, or see in advance that they must belong to a single subject.
     From: Gottfried Leibniz (New Essays on Human Understanding [1704], 2.31)
     A reaction: See Idea 12975 for imperfect ideas of things. Obviously the idea is that perfect knowledge will converge on a single definition, which will pinpoint the essence of a thing, and then all explanations will flow. A nice addition to the Aristotelian view.
One essence can be expressed by several definitions [Leibniz]
     Full Idea: Although a thing has only one essence, this can be expressed by several definitions.
     From: Gottfried Leibniz (New Essays on Human Understanding [1704], 3.03)
     A reaction: See Idea 12976 and Idea 12977 for a view which seems to conflict with this. He seemed to imply that once you identify the essence, the definitions converge, with multiple definitions being symptomatic of imperfect ideas of things.
Real definitions, unlike nominal definitions, display possibilities [Leibniz]
     Full Idea: The real definition displays the possibility of the definiendum, and the nominal definition does not.
     From: Gottfried Leibniz (New Essays on Human Understanding [1704], 3.03)
     A reaction: This, I take it, is because the real definition indicates the actual powers of the thing, and not just the superficial characteristics. Is knowledge of powers identical with knowledge of possibilities?
Definitions can only be real if the item is possible [Leibniz]
     Full Idea: Definitions to my mind are real, when one knows that the thing defined is possible; otherwise they are only nominal, and one must not rely on them.
     From: Gottfried Leibniz (Letters to Antoine Arnauld [1686], 1686.07.4/14 XI)
     A reaction: It is interesting that things do not have to actual to have real definitions. For Leibniz, what is possible will exist in the mind of God. For me what is possible will exist in the potentialities of the powers of what is actual.
A real definition gives all the properties that constitute an identity [Molnar]
     Full Idea: A real definition expresses the sum of the properties that constitute the identity of the thing defined.
     From: George Molnar (Powers [1998], 1.4.4)
     A reaction: This is a standard modern view among modern essentialists, and one which I believe can come into question. It seems to miss out the fact that an essence will also explain the possible functions and behaviours of a thing. Explanation seems basic.
Chemists aim at real definition of things; lexicographers aim at nominal definition of usage [Gupta]
     Full Idea: The chemist aims at real definition, whereas the lexicographer aims at nominal definition. ...Perhaps real definitions investigate the thing denoted, and nominal definitions investigate meaning and use.
     From: Anil Gupta (Definitions [2008], 1.1)
     A reaction: Very helpful. I really think we should talk much more about the neglected chemists when we discuss science. Theirs is the single most successful branch of science, the paradigm case of what the whole enterprise aims at.
Definitions formed an abstract hierarchy for Aristotle, as sets do for us [Fine,K]
     Full Idea: For us it is sets which constitute the most natural example of a hierarchical structure within the abstract realm; but for Aristotle it would have been definitions, via their natural division into genus and differentia.
     From: Kit Fine (Aristotle on Matter [1992], §1 n4)
     A reaction: I suppose everyone who thinks about reality in abstraction ends up with a hierarchy. Compare the hierarchy of angelic hosts, or Greek gods. Could we get back to the Aristotelian view, instead of sets, which are out of control at the top end?
Modern philosophy has largely abandoned real definitions, apart from sortals [Fine,K]
     Full Idea: In modern analytic philosophy we find that, as a result of sustained empiricist critique, the idea of real definition has been more or less given up (unless it be taken to be vestigially present in the notion of a sortal).
     From: Kit Fine (Essence and Modality [1994], p. 3)
     A reaction: The account of essences as falling under sortals (roughly, categorising terms) is associated with David Wiggins. Kit Fine is in the business of reviving Aristotelian real definitions, as are fans of scientific essentialism (see under 'Nature').
Maybe two objects might require simultaneous real definitions, as with two simultaneous terms [Fine,K]
     Full Idea: In Wooster as the witless bachelor and Jeeves as the crafty manservant, and one valet to the other, we will have the counterpart, within the framework of real definition, to the simultaneous definition of two terms.
     From: Kit Fine (Ontological Dependence [1995], III)
     A reaction: This is wonderful grist to the mill of scientific essentialism, which endeavours to produce an understanding through explanation of the complex interactions of nature.
'Real' definitions give the essential properties of things under a concept [Mautner]
     Full Idea: A 'real definition' (as opposed to a linguistic one) is a statement which gives the essential properties of the things to which a given concept applies.
     From: Thomas Mautner (Penguin Dictionary of Philosophy [1996], 'definition')
     A reaction: This is often seen as old-fashioned, Aristotelian, and impossible to achieve, but I like it and aspire to it. One can hardly be precise about which properties are 'essential' to something, but there are clear cases. Your 'gold' had better not be brass.
A sentence may simultaneously define a term, and also assert a fact [Boghossian]
     Full Idea: It doesn't follow from the fact that a given sentence is being used to implicitly define one of its ingredient terms, that it is not a factual statement. 'This stick is a meter long at t' may define an ingredient terms and express something factual.
     From: Paul Boghossian (Analyticity Reconsidered [1996], §III)
     A reaction: This looks like a rather good point, but it is tied in with a difficulty about definition, which is deciding which sentences are using a term, and which ones are defining it. If I say 'this stick in Paris is a meter long', I'm not defining it.
A successful Aristotelian 'definition' is what sciences produces after an investigation [Koslicki]
     Full Idea: My current use of the Aristotelian term 'definition' is intended to correspond to what is typically accessible to a scientist only at the end of a successful investigation into the nature of a particular phenomenon.
     From: Kathrin Koslicki (Essence, Necessity and Explanation [2012], 13.3.1)
     A reaction: It is crucial to understand that Aristotle's definitions could be several hundred pages long. It has nothing to do with dictionary definitions. He proposes 'nominal' and 'real' definitions.
Real definitions don't just single out a thing; they must also explain its essence [Koslicki]
     Full Idea: A statement expressing a real definition must also accomplish more than simply to offer two different ways of singling out the same entity, since the definiens must also be explanatory of the essential nature of the definiendum.
     From: Kathrin Koslicki (Varieties of Ontological Dependence [2012], 7.4)
     A reaction: This is why Aristotelian definitions are not just short lexicographical definitions, but may be quite length. Effectively, a definition IS an explanation.
'Nominal' definitions identify things, but fail to give their essence [Jones,J-E]
     Full Idea: In the Aristotelian tradition, a 'nominal' definition is a pseudo-definition that identifies the members of the species or genus, but fails to capture the essence, e.g. 'man is the featherless biped'.
     From: Jan-Erik Jones (Real Essence [2012], §2)
     A reaction: You can 'individuate' an object as 'the only object in that drawer', while revealing nothing about it. So what must a definition do, in addition to picking something out uniquely?