It was not long ago discussed by Ramsey in the form
"What is the difference between a particular and a universal?", and more recently by Ayer in the form
"What is the difference between an individual and a property?"
Ramsey decides that there was no ultimate difference.
But perhaps Ramsey set the standard for an ultimate difference higher than we should wish to, or drew it from a theory we no longer wish to hold.
Ayer, after some interesting suggestions, changed the subject, and discussed instead two other questions:
viz.,
what is the difference in function between indicator words and predicates, and could we in principle say what we want to say without using the former ?
It may be that the original question is made easier to start on, and more difficult to settle, by an initial failure to make even fairly clear what types or classes of things are to be included in the two general categories between which a satisfying difference is sought.
The words of the questions I quoted are not very helpful.
A universal, or a universalium (Aristotle's 'katholou") may be a quality or a relation.
(Strawson is much indebted to Grice for his criticisms).
To the second question Ayer's answer was affirmative ; and, things being as they are, this is no doubt correct as a matter of what is theoretically practicable.
Ayer also acknowledges
a) that in actual practice we could scarcely dispense with indicator words, and
b) that the attempt to do so would always involve a theoretical failure to individuate, since no elaboration of predicates rules out the theoretical possibility of reduplication.
But I doubt if the original question can be answered unless we take these two facts more seriously than he does.
But if we take, e. g., the expressions "quality," "relation" and "property" in their current uses, much that we should no doubt wish to include on the side of the "general," as opposed to the "particular," would be left out.
And if we do not take them in their current uses, it is not clear how we are to take them. Aristotelian usage? (I am indebted to Grice for the suggestion).
Thus snow, gold and clothing is, neither a property ; nor is man, nor any other species ; nor is chess nor furniture ; nor is the Union Jack—by which I mean, not the tattered specimen the porter keeps in a drawer, but the flag designed in the 19th century, examples of which are taken from drawers by porters and hung from windows.
But all these are things which we might well wish to classify with properties correctly so-called, like inflammability, or with qualities correctly so-called, like prudence, when we contrast these latter with individuals or particulars.
For there are individual flakes or drifts or falls of snow, pieces of gold, articles of clothing or furniture, games of chess, members of species ; and there are hundreds of Union Jacks.
These are all (are they not?) particular instances of the general things named in their names.
Sometimes the unlikeness of these general things to properties or qualities correctly so-called is masked by the introduction of expresssions like " being (a piece of) gold," "being snow," " being a man," " being a Union Jack," " being a chair," " being a game of chess "—phrases like these being said to name properties.
Now such expressions no doubt have a participial use ; and some (e.g., " being a man ") may have a use as noun-phrases, as singular terms.
But it is dubious whether many of them have a use as singular terms ; and it is dubious whether any of them can be regarded as names of properties.
And however we resolve these doubts in different cases, the following dilemma arises in each.
Either these verbal nouns (where they are nouns) have the same use as the general names they incorporate—and in that case they may as well be discarded in favour of those general names, which are more familiar, and about the use of which we are consequently less liable to be misled ; or they have a different use from those general names—and in this case
(But a game of chess may be something which itself has instances.)
we still have on our hands, to be differentiated, like properties correctly so-called, from particulars, the general things designated by those familiar general names.
This initial unclarity about the limits of the two great categories of general and particular shows itself also in that arbitrary narrowing of the field which must be presumed to occur whenever certain answers to our question seem plausible.
I shall consider again some of these answers, which were dismissed by Ayer or Ramsey or both, not so much on the ground that they thought them false as on the ground that they did not think them fundamental.
There is, for example, the suggestion that a general, unlike a particular, thing cannot be perceived by means of the senses ; and this seems most plausible if one is thinking of the things designated by certain abstract nouns.
It is not with the eyes that one is said to see hope. But one can quite literally smell blood or bacon, watch cricket, hear music or thunder ; and there are, on the other hand, certain particulars which it makes dubious sense to say one perceives. Then there is the suggestion that general, unlike particular, things, can be in several places at once.
There can be influenza in London as well as in Birmingham, and gold in Australia as well as in Africa.
But then so can many particulars be scattered over the surface of the table or the globe.
Moreover, it makes dubious sense to say of some general things (e.g., solubility) that they, are in any place, let alone in many ; and equally dubious sense to say of some particular things (a sudden thought, a mental image, the constitution of France) that they have a particular spatial location. It may be said that I have missed the point of both these theories ; that, first, when we say we perceive general things, what we really perceive is individual instances of them, not the general things themselves ; and, second, to say that general things can be in several places at once is to say that they may have different instances, differently located ; whereas it makes no sense to speak of different instances of individuals.
But so to explain these theories is to give them up. It is to fall back on saying that general things may have instances, and individual instances of general things may not.
This is, perhaps, an unexceptionable statement of the general distinction between the two categories, but scarcely seems to count as an explanation of it.
A third suggestion is that individual things, unlike general things, have dates or histories.
But similar objections apply to this.
We may speak of the history of dress or engineering, the origins of civilization, the invention of golf and the evolution of man.
This theory, like the others (when taken at their face value), may draw a logically interesting distinction ; but, like them, does not draw one that coincides with the categorial line between particular and general. A doctrine which might appear more promising, because more general, than these, is that individuals can function in propositions only as subjects, never as predicates ; whereas general things can function as both. But it is not clear what this doctrine amounts to.
Suppose, first, it is a grammatical point. Then if it says that the names of individuals never have adjectival or verbal forms, whereas names of general things do, it is false. If it says that individual names never form parts of grammatical predicates, or alternatively, never stand by themselves after the word " is " in a grammatical predicate, it is equally false. In any case, a grammatical point could scarcely be fundamental, since it is easy to imagine the elimination of those distinctions upon which such points must rely, in favour of the device of merely coupling names of appropriate types, in any order, in a singular sentence. We should not, by so doing, eliminate the category-distinction. For we might imagine changing the language once more, requiring that our names should stand on one side or the other of the phrase " is an instance of," and then simply distinguishing the individual names as those that could never stand on the right of this phrase.
( 4 Ramsey seems to suggest that this would simply be to manufacture an empty verbal distinction. (Cf. Foundations of Mathematics, pp\ 132-133).
But it would not. For it would not be an arbitrary matter to decide which names to put on which side of the coupling phrase)
So I think we must conclude that the point misleadingly made in the languages of grammar is simply once more the point that individuals, unlike general things, cannot have instances. To say that general things, unlike individuals, can be predicated of other things, is simply to paraphrase this ; and neither expression seems more perspicuous than the other.
But will the word " instance " itself really bear the* weight of this distinction ?
Of course, as a philosopher's word, understood in terms of that distinction, it cannot fail to bear it ; but then it ceases to explain the distinction for us.
If we ask what expressions we actually use to refer to or describe an individual thing as an instance of a general thing, we find that they are many ; and that perhaps none of them is appropriate in every case. They include : " a case of," "a n example of," " a specimen of," " a member of," " a piece of," " a quantity of," " a copy of," " a performance of," " a game of," " an article of," and so on. Though each can be followed by the name of a general thing, many can also be followed by expressions we should hesitate to regard as the names of general things. This is true of the phrase " an instance of" itself. We may speak of a signal instance of generosity ; but we may also speak of a signal instance of Smith's generosity. Similarly we may speak not only of a piece of gold and an article of clothing, but of a piece of Smith's gold and an article of Smith's clothing. So if we seek to draw our distinction in terms of the words actually used to play the part of the philosopher's word " instance "—including the word " instance " itself—then it will not be enough to say that general things may have instances. For so may non-general things. The point here may be put roughly as follows. We are tempted to explain the distinction between two types of things, Tx and T2, by means of a certain relation R ; by saying, that is, that only things belonging to T2 can appear as the second term of this relation, whereas both things belonging to T2 and things belonging to Tx can appear as its first term. R is something like, but more general than, is characterized by or is a member of or the converse of is predicated of. But then it appears that we really have no notion of R except one which is useless for explanatory purposes since it is itself to be explained in terms of the difference between Tx and T2 ; this is what I called the philosopher's notion of " an instance of."
What we have instead is a lot of notions which are either too restricted to serve our purpose (e.g., " has the property of"), or fail to be restricted in precisely the way in which we want them to be, or both. As a member of this set of notions, preeminent for its abstract character, we may take the logician's idea of class-membership. The difficulty is, roughly, that we can form closed classes on what principle we please ; we could count almost any particular we are likely to mention as such a class, and hence as the second term of our relation. (These remarks are very rough and schematic ; but they serve, I hope, to make the point in a general form). Consequently, we shall have to give up the idea of explaining the difference between the particular and the general in terms of such a relation.
This will not lead us, as it perhaps led Ramsey, to despise the philosopher's notion of an instance, and to think that there is nothing in it ; for it is easy enough to teach anyone the application of it, without precise explanations. But it will lead us to look further for such explanations. 4. To begin with, I want to draw a rough distinction between three classes of nouns, all of which would traditionally be regarded either as themselves the names of universals (general things) or—in the case of the nouns of group (2)'—as closely linked to such names. The distinctions are indicated only by examples ; and the three classes are by no means exhaustive of the field. But this does not matter for my purpose.
(1) Examples of the first class are such partitive nouns as "gold," "snow," "water," "jam," "music." These I shall call material-names, and what they name, materials.''
(2) Examples of the second are certain articulative nouns such as " (a) man," " (an) apple," " (a) cat."
These I shall call substance-names, and what they apply to, substances.
Examples of the third are such abstract nouns as "redness," (or "red"), "roundness," "anger," "wisdom."
These I shall call quality- or property-names, and what they name, qualities or properties.'7 These three classes of nouns may be compared and contrasted with one another in a number of ways. But the contrast on which I wish to lay most emphasis is (i) The contrast between the nouns of group (3) and those of groups (1) and (2).
The nouns of group (3) are the most sophisticated and the most dispensable. They are derived from adjectives and the general things they name usually enter our talk by way of the adjectives from which their names are derived. When we consider the things which philosophers are prepared to count as individual instances of these general things, we find a considerable latitude in the categories of the things to which these instances may belong. Thus an instance of wisdom may be a man, a remark or an action. An instance of the colour red may be a material thing like a pillar-box, an event like a sunset, or a mental thing like an image.
A word, a gesture, an expression, a man may all be instances of anger. In contrast, unsystematic ambiguities aside, there is no latitude at all about what category of thing can be. an individual instance of a cat or an apple. There is some latitude, but one would often hesitate to call it a category-latitude, about what can be an individual instance of the general things named by the nouns of group (1). An instance of gold may be a vein, a piece or a quantity of gold ; an instance of snow may be a drift, an expanse, a piece, and even a fall, of snow. (ii) Next I want to emphasise a respect in which the nouns of group (2) differ from those of groups (1) and (3). Philosophers may speak of "a n individual (particular) ' The terminology, evidently, is not to be taken too seriously. Anger is a state, not a property or quality.
instance (example, specimen)of <£," where " 4> " is replaced by a noun from any of these three groups. Suppose the noun is drawn from group (2). Then we have such phrases as " an instance of a horse " or "a n instance of an apple." It is to be noticed that what follows the expression " an instance of " is a phrase which can and does by itself function as an indefinite designation of an individual instance. (An instance of a horse is the same as a horse.) This is not the case if the nouns are drawn from groups (1) or (3). (Gold is not the same as a piece of gold.) It seems as if, when we say that x is an instance ofy, then whenj is such that there is no choice about the sort of thing we can count as an instance of it, we feel no need of a true general-thing name for y, i.e., of a name differing from an indefinite designation of an individual instance ofy. (It is true that we have the expressions " the horse," " the apple," etc., names of species or kinds, obvious collectors of homogeneous individuals ; but these follow less naturally after the expression "an instance of" than does the phrase containing the indefinite article).
Plato (or is it Antisthenes?) who feels this difference, and tries to blur (or enhance) it with the invention of such expressions as "horseness" (cf " being a horse ").
ἱππότης, horse-nature, the concept of horse
(Antisth. et Pl. ap. Simp.in Cat.208.30,32, Sch.AristId.p.167F).
But it should rather be treated as a clue until proved an anomaly.
Finally, I want to note the existence of a special class of individual instances of general things whose names belong to group (3).
The simplest, though not the only recipe, for forming the names of members of this class is as follows : in the formula " the . . . of . . .," fill the first gap with the property-name in question and the second gap with the definite designation of a suitable individual.
Thus we may speak of the wisdom of Socrates as an instance of wisdom ; of the redness of Smith's face as an instance of redness ; and we may also speak of Jones' present mental state as an instance of anger.
This class of individual instances of properties, or property-like things, will include the " particular qualities " which Stout defended.
And an analogy may be found between referring to a horse as " an instance of a horse " and referring to Jones' present stage of anger as " an instance of anger."
Next, I want to make some general, and still propaedeutic, remarks about the notion of an individual or particular.
The idea of an individual is the idea of an individual instance of something general.
There is no such thing as a pure particular. (This truth is too old to need the support of elaboration.)
The idea of an individual instance of j> is the idea of something which we are able in principle (a) to distinguish from other instances of <}> ; and (b) to recognise as the same instance at different times (where this notion is applicable). So, to have the idea of a particular instance of
" is a substancename (4(ii)). It should once more be noted that these are
the cases where we do not find a true name of a general
thing following the phrase " an instance of," but instead
an expression which can by itself function as an indefinite
designation of an individual instance (e.g., " a horse ").
When it has been said that a particular must be an instance of something general, and that there must be criteria of distinctness and (where applicable) of identity for individual instances of a general thing, something of central importance still remains unsaid. In giving the relevant criteria'—or sets of criteria—for individual instances of a certain general thing, we do not indicate how such particulars are brought into our discourse. Nor do we bring a particular into our discourse by mentioning these criteria. (To mention them is still to talk in general.)
We bring a particular into our discourse only when we determine, select, a point of application for such criteria, only when we mention, refer to, something to which these criteria are to be applied ; and no theory of particulars can be adequate which does not take account of the means by which we determine such a point of application as a point of application for these criteria.
In the rest of this paper I shall try to do two things.
First, I shall try to show how, in the case of certain kinds of particulars (particular instances of certain kinds of general things), the notion of a particular may be seen as something logically complex in relation to other notions (a kind of compound of these notions). That is, I shall try to produce a partial explanation (analysis) of the notion of an individual instance, for certain cases ; and then I shall try. to show how this notion, as explained for these cases, may be used in the explanation of the notion of individual instances of other sorts of general things, and in the explanation of the notions of those other types of general things themselves. So in this part of the paper (sections 6-9), no general account is offered of the distinction we are concerned with. The procedure is essentially one of indicating, step by step, how certain types of notion can be seen as depending upon others ; and it makes no claim at all to completeness. Second (section 12), this procedure is found to suggest a possible general account of the distinction we are concerned with ; though the acceptability or otherwise of this general account seems to be independent of that of the step-by-step schema of explanation. Now it might seem that the difficulty of finding an explanation of the notion of an individual instance arises from the fact that the category distinction between general and individual is so fundamental that there is nothing logically simpler, or more fundamental, in terms of which this notion could be explained.
But I think this view can be challenged for a certain range of important cases, which can then perhaps serve as the basis for the explanation of others. To challenge it successfully, we have to envisage the possibility of making statements which (a) do not make use of the notion of individual instances, and (b) do not presuppose the existence of statements which do make use of this notion. The second condition may be held to rule out general statements ; for though many general statements make no direct mention of individuals, they have often and plausibly been held in some sense to presuppose the existence of statements which do.
So what we have to consider is the possibility of singular statements which make no mention of (i.e., contain no names for, or other expressions definitely or indefinitely referring to) individual instances of general things.
Now there certainly does exist, in ordinary use, a range of empirical singular statements answering to this description.
I suggest, as examples, the following :—
It is (has been) raining
Music can be heard in the distance
Snow is falling
There is gold here
There is water here.
All these sentences contain either the material-name of a general thing (" music," " snow ") or a corresponding verb ; but none contains any expression which can be construed as serving to make a definite or indefinite mention of individual instances of those general things (i.e., falls or drops of rain, pieces of gold, pools of water and so on).
Of course, when these sentences are used, the combination of the circumstances of their use with the tense of the verb and the demonstrative adverbs, if any, which they contain, provides an indication of the incidence of the general thing in question. Such an indication must be provided somehow, if empirical singular statements are to be made at all.
But it is important that it can be provided by means of utterance-centred indications which do not include noun expressions referring definitely or indefinitely to individual instances.
Such sentences as these do not bring particulars into our discourse.
Languages imagined on the model of such sentences are sometimes called " property-location " languages.
But I think the word " property " is objectionable here because
a) the general things which figure in my examples are not properties, and
b) the idea of a property belongs, with the idea of an individual instance itself, to a level of logical complexity we are trying to get below.
So I propose to substitute the less philosophically committed word " feature " ; and to speak of feature-placing sentences.
Though feature-placing sentences do not introduce particulars into our discourse, they provide the materials for this introduction.
Suppose we compare a feature-placing sentence (" There is snow here ") with a phrase (" This (patch of) snow ") in the use of which an individual instance of the feature is mentioned. It seems possible, in this case, to regard the notion of the individual instance as something logically complex in relation to the two simpler notions of the feature and of placing.
"The logical complexity may be brought out in the following way. In making the featureplacing statement, we utter a completed sentence without mentioning individuals. If we merely mention the individual without going on to say anything about it, we fail to utter a completed sentence ; yet what the feature-placing sentence does explicitly is, in a sense, implicit in this mere mention.
So, as the basic step in an explanatory schema, we may regard the notion of a particular instance of certain sorts of general things as a kind of logical compound of the simpler notions of a feature and of placing. But what about the criteria of distinctness and identity which were said in general to be necessary to the notion of an individual instance of a general thing ? The basis for the criteria of distinctness can already be introduced at the feature-placing level, without yet introducing particulars. For where we can say " There is snow here " or " There is gold here," we can also, perhaps, more exactly, though not more correctly, say " There is snow (gold) here—and here— and here." And when we can say " It snowed to-day," we can also, perhaps, more exactly, but not more correctly, say " It snowed twice to-day."
The considerations which determine multiplicity of placing become, when we introduce particulars, the criteria for distinguishing this patch of snow from that, or the first fall of snow from the second. Of criteria of identity I shall say more in general later. It might be objected that it is absurd to speak of an imagined transition from feature-placing sentences to substantival expressions definitely designating particular instances of features as the introduction of particulars ; that it is absurd to represent this imagined transition as part of a possible analysis of the notion of a particular instance, even for these simple cases of material-names which seem the most favourable ; and that at most what is achieved is the indication of a possible way of looking at certain designations of certain particulars.
For are not the particulars as much a relevant part of the situation in which a featureplacing sentence is employed as they are of a situation in which a substantival particular-designation is employed ? To this I would reply by asking what philosophical question there would be about particulars if we did not designate them, could not make lists of them, did not predicate qualities of them and so on. What we have to explain is a certain mode of speech.
When we turn from material-names to substancenames, the attempt to provide an analogous explanation of the notion of an individual instance seems much harder. But though it is harder, it is perhaps worth making ; for if it succeeds, we may find we have then an adequate basis for the explanation of the notion of an individual instance in other cases, and for the explanation of further kinds of general things. In order for the attempt to succeed, we must be able to envisage a situation in which, instead of operating with the notion of an individual instance of a cat or an apple, we operate with the notions of a corresponding feature and of placing.
Ordinary language does not seem to provide us, in these cases, with feature-placing sentences.
And it might be argued that the idea of such sentences was, in these cases, absurd.
For (1) it might be pointed out that an all-important difference between such things as snow and such things as cats lay in the fact that different instances of snow are, in a sense, indefinitely additive, can be counted together as one instance of snow ; while this is not true in the case of instances of cats ; and it might be suggested that herein lay a reason for the possibility of feature-placing sentences in the case of snow and for their impossibility in the case of cats.
And (2) it might be added that we have no name for a general thing which could count as the required feature in the case of cats. It is true that we speak of the cat in general ; but " the cat " ranks as a species-name, and the notion of a species as surely presupposes the notion of individual members as the notion of a property involves that of individual things to which the property belongs or might belong. It is also true that we may speak of an instance (specimen) of a cat, as we may speak of an instance of gold ; but here what follows the phrase " an instance of" is not, as " gold " is, a generalthing name which could figure in a merely feature-placing sentence, but an expression which also serves as an indefinite designation of an individual.
Does not all this strongly suggest that there could be no concept of the " cat-feature " such as would be required for the analysis to work, that any general idea of cat must be the idea of a cat, i.e., must involve criteria of identity 'ind distinctness for cats as individuals and hence the notion of an individual instance ? These objections have great force and importance ; but I do not think them decisive. For they do not show that it is logically absurd to suppose that we might recognise the presence of cat or signs of the past or future presence of cat, without ever having occasion to distinguish one cat from another as the cat on the left, or identify a cat as ours or as Felix.
The second argument merely reminds us that the resources of our language are such that on any actual occasion of this kind we in fact use, not a partitive noun, but the indefinite forms (" cats " or " a cat") of the articulative noun. But this fact can be explained in a way consistent with the advocated analysis (see section 10). Nevertheless, these arguments show something. The point about the species-name, for example, is sound ; the notion of a species, like that of a property, belongs to a level of logical complexity we are trying to get below. Second, and more immediately important, the first argument shows that if there is to be a general concept of the cat-feature, corresponding in the required way to the notion of an individual instance, it must already include in itself the basis for the criteria of distinctness which we apply to individual cats. (Roughly, the idea of cat, unlike that of snow, 9 Cf. Price, Thinking and Experience, pp. 40-41, on identity of individuals and of characteristics.
would include the idea of a characteristic shape). But to concede this is not to concede the impossibility of the analysis. It is worth adding that sometimes we do find verbal indications of our use of feature-concepts such as those we are trying to envisage ; as, e.g., when we speak of " smelling cat " or " hunting lion," using the noun in the singular without the article. There might seem to exist a more general objection to this whole procedure.
For it seems that it would always be possible in practice to paraphrase a given feature-placing sentence in use, by means of a sentence incorporating indefinite designations of particular instances ; e.g., " There is gold here " by " There is a quantity of gold here " ; " Snow has fallen twice " by " There have been two falls of snow " ; " There is snow here—and here " by " There are patches (expanses) of snow here and here " ; and so on. And if sentences incorporating definite or indefinite designations of particular instances bring particulars into our discourse ; and if statements made by the use of featureplacing sentences are equivalent to statements made by the use of sentences incorporating indefinite designations of particular instances ; then do not feature-placing sentences themselves bring particulars into our discourse ? But this argument can be turned in favour of the explanation it is directed against.
Suppose there is a statement S made by means of a feature-placing sentence ; and an equivalent statement S' made by means of a sentence incorporating an indefinite particular-designation ; and a statement T made by means of a sentence incorporating a definite designation of the particular indefinitely designated in S'. Now only if a language admits of statements like T can it admit of statements correctly described as I have described S'. (There are no indefinite designations of particulars where there are no definite designations of particulars.) But a language might admit of statements like S without admitting of statements like T. So the existence of statements like S', in a language which admits of both statements like S and statements like T, is not destructive of the analysis, but is a proof of its correctness.
If the argument so far is acceptable, then at least in the case of some materials and some substances, we can regard the notion of an individual instance as partially explained in terms of the logical composition of the two notions of a feature and of placing.
When we turn to properties and qualities, we may make use of a different kind of explanation which is also, in a sense, the completion of the first kind. I shall not, that is to say, try to explain the notion of individual instances of anger or wisdom or red in terms of the logical composition of a feature, such as anger or red, and placing. But nor shall I maintain that it would be wrong or impossible to do so. We might think of such general things as anger (or red) not primarily as qualities, properties, states or conditions of persons or things, but primarily as instantiated in, say, situations (or patches) which acquired their status as individuals from just such a logical composition.
But though this is how we might think, it is not, for the most part, how we do think. It is natural, rather, to regard those general things which are properly called qualities, conditions, etc., as belonging at least to the same level of logical complexity as the idea of individual instances of the kinds we have so far been concerned with ; to regard them, that is, as feature-/?^ things, the incidence of which, however, is primarily indicated, not by placing, but by their ascription to individual instances of material or substantial features the incidence of which is primarily indicated by placing.10 We have seen that the notion of an individual instance of some materials and substances can be regarded as a logical compound of the notions of a feature and of placing. We have now to see the ascription of a quality (etc.) to such an individual as an operation analogous to the placing of a feature. Indeed, we may find in the possibility of this operation the point—or one important point—of that logical composition which yields us the particular. The individual instance of the simply placeable feature emerges as a possible locationw These remarks, of course, apply only to some of the things correctly called properties, states, qualities, etc. 2
point for general things other than the feature of which it is primarily an instance, and hence as also an individual instance of these general things, its properties or qualities or states. One might exaggeratedly say : the point of having the idea of individual instances of material or substantial features is that they may be represented as individual instances of property-like features. The individuals are distinguished as individuals in order to be contrasted and compared. Other notions call for other treatment. I consider two more. (a)
I mentioned, at 4 (iii), a rather special class of individual instances of properties or property-like things. We form the notion of such an instance when, for example, we speak not of a man or an action as an instance of wisdom or anger, but of the wisdom of Socrates as an individual (a case of wisdom) or of Jones' present mental state as an individual instance of anger.
Here the notion of the individual instance can be seen as a new kind of logical compound, namely, a compound which includes as elements both the notion of the general thing (property) in question and that of the material or substantial individual which is an instance of it ; it may sometimes include a further element of temporal placing (cf. " his present state of anger "). (b) Instances of events, processes and changes I have so far scarcely mentioned.
Most of our most familiar words for happenings strike us essentially as names for the actions and undergoings of individual instances of material or substantial features. But there is a difference between these happening words and quality or state-words. A wise man is an instance of wisdom, but a dead or dying man is not an instance of death. Only a death is that. As regards such happening-words as these, then, we have to see the idea of an individual instance as reached by a kind of logical composition analogous to that considered in the paragraph immediately above : an individual instance of a material or substantial feature is an element in the compound.
But these, though perhaps the most important, are not the only kinds of happening-words.
9. The general form of these explanations may be roughly indicated as follows. The notion of placing a feature is taken as basic, as consisting of the logically simplest elements with which we are to operate. It is pointed out that neither of these elements involves the notion of an individual instance, nor therefore the notions of certain types of general things, such as properties and species ; and it is shown that the idea of operating solely with these simplest elements can be made intelligible for certain cases. (Features in fact of course belong to the class of general things ; but so long as we remain at the feature-placing level, they cannot be assigned to it ; for there is nothing to contrast the general with.) From this basis we proceed by composition and analogy.
The designations of individual instances of (some) material and substantial features are first introduced, as expressions, not themselves complete sentences, which include placingindications ; and, complementarily, certain types of general things (e.g., properties and types of happening) are introduced as items the designations of which do not include placing-indications and which are ascribed to material or substantial individuals. The ascription of such a thing as a property to a substantial individual is represented simply as an operation analogous to the placing of a feature ; so no circularity attends the word " ascription." Individuals of certain other types {e.g., events happening to substances, states of substances and " particularised" qualities) are then introduced as the designata of expressions which include the designations of individuals of earlier types, and hence indirectly include the notion of placing.
There are many types of individual and of general thing besides these here considered. Some may admit of analogous treatment ; and it might be possible to introduce others, on the basis already provided, by other methods of construction and explanation. But every introduction of a particular, in terms of such a schema, will either directly contain the notion of placing or will preserve, by way of individuals already introduced, the original link with this notion.
Of course the value of this suggestion, as it stands, is small. For the notion of an individual instance extends itself indefinitely, by way of far more complicated connexions than I have so far indicated ; and the limits of plausibility for the kinds of construction-procedure I have used would, no doubt, soon be reached, if they are not already overpassed. Nevertheless, I think this sketch of a procedure has certain merits :
Some of the difficulties which attend any attempt to elucidate the category-distinction between the particular and the general arise from the fact that these two classes include so many different category-distinctions within themselves.
This fact creates a dilemma for the theorist of the distinction. On the one hand, he is tempted, in a way illustrated at the beginning of this paper, into drawing distinctions which indeed separate one or more sub-categories of one class from one or more sub-categories of the other, but which fail to yield the desired result if applied over the whole field.
Or, on the other hand, in the effort to escape from this domination by irrelevant category differences, he is tempted by the prospect of a purely formal distinction, drawing for this purpose on the terms and concepts of grammar or of formal logic.
But distinctions so drawn can only seem to succeed by forfeiting their formal character and silently incorporating the problematic category-distinction.
The present procedure offers at least a hope of escape from this difficulty. For it fully allows for the differences between types of general thing and of individual ; and instead of producing one single explanation, the same for every case, it offers a serial method of explaining later types of general or particular things on the basis of earlier ones, while preserving a continuous general differentiation between the two major categories in the course of the explanation.
Too much must not be claimed for the suggested procedure, however • in particular, it must not be thought that it has been so described as to provide a criterion for the distinction we are concerned with.
Another characteristic of the schema of explanation is that it accords a central place to the notion of an individual instance of certain kinds of general things, viz., of material and substantial features. This (see section 8) is not an essential characteristic ; it could be modified. But there is reason to think that it corresponds to our actual way of thinking ; that these individuals are the " basic particulars." Why this should be so, and whether it might not be otherwise, are questions which I shall not now consider.
Finally, while not itself providing a criterion of general and particular, the schema points the way to a possible general distinction which might be defensible even if the procedure which suggests it should prove unsatisfactory. This general distinction I shall outline in section 12. Before I do so, some further points remain to be considered.
Something further must first be said on the subject of criteria of identity for individual instances of a general thing. We saw (5 (2)) how in many cases the question of the criteria of distinctness and identity of an individual instance of a general thing was incompletely determined when the general thing was named. This "was particularly evident in«the case of some properties and was evident also in the case of materials. Where substance-names were concerned, however, this indeterminateness seemed not to exist ; when the name was given, the criteria were fixed.
And this was connected with the fact that in these cases there seemed to exist no true general-thing name, apart from expressions which ranked as species-names and obviously presupposed certain definite criteria of identity for individual members. As far as criteria of distinctness are concerned, this raises no particular difficulty.
We saw, for example, how the idea of a simply placeable feature might include—might indeed be—the idea of a characteristic shape, and in this way provide a basis for criteria of distinctness for individual instances of the feature. But it is not so easy to account for the apparent determinateness of criteria of identity. The explanatory schema advanced required that we should theoretically be able to form concepts of some substance-features which were logically prior to, and independent of, the corresponding concepts of an individual instance of such features ; and this requirement seems to clash with the apparent determinateness of the criteria of identity for such individuals. A parallel answer to that given in the case of criteria of distinctness is theoretically available, but is unattractively unplausible.
If we reject this answer, and cannot find an alternative, then we must at least radically revise, though in a not unfamiliar direction, the basis of the explanatory schema. (The difficulty is essentially a more specific form of that encountered already in section 7). I think, however, that an acceptable alternative can be found. For in all cases where a feature-concept can be assumed to be possible, the criteria of identity (and of distinctness) for an instance of the general thing in question •—or the sets of such criteria, where there is more than one set—can be seen as determined by a combination of factors, viz., the nature of the feature itself, the ways in which the feature empirically manifests itself in the world, and—to adopt a possibly misleading mode of expression—the kind of incentives11 that exist for having a notion of an individual instance of the feature in question.
The rele\4ance of this third factor even, perhaps, gives us the right to say that there is something arbitrary about the criteria we adopt, something which, given the other two factors, is—in at any rate a stretched sense—a matter of choice. In extreme cases this is obvious. Even those who had witnessed the whole of the affair under discussion might, for example, give varying answers to such a question as : Is this the same quarrel going on now as was going on when I left ? The answer we choose may depend on just what distinctions we are interested in ; and one can imagine many situations 11
What I mean by " incentives " here may be illustrated from the con- venience of the institution of property. Suppose there is a general feature,, which human beings wish to make use of. Even if there is enough ^ for all, friction may be avoided if criteria are used for distinguishing my ^ from yours. (" Mine " is indeed one of the earliest individuating words used by
children.)
for this example, and many different things which might influence us. There may, on the other hand, be very many cases of features where the adoption of a certain particular set of criteria of identity (and distinctness) for their instances is so utterly natural that it would seem to be stretching the phrase " matter of choice " intolerably to apply it to them.
But, even in these cases, the naturalness may still be seen as depending on the combination of factors I mentioned ; and, if we bear this in mind, we can sometimes imagine the possibility of alternatives. (Here is a question which might with advantage be explored for many different types of case.) It seems reasonable to view substantial features as cases of this kind. If this view, is acceptable, we can find in it an explanation of that difference between substance-names and certain other true general-thing names to which I have several times referred.
Given a true general-thing name, like " gold " or " wisdom," the question of the criteria of identity of its instances cannot be answered until the kind of instance is specified, by such a phrase as " apiece of gold " or " a wise action".
But where one set of criteria of identity is peculiarly dominant, its adoption peculiarly compelling, we find no such non-committal general name in current, adult, unsophisticated use.
All that we might wish to do with it, we can equally well do without it, by the use of the indefinite singular or plural forms of the ordinary substancename (e.g., " a horse " or " horses "). 11.
It is, perhaps, necessary to guard briefly against a misunderstanding. Of course, I am not denying that we can very well use individual-designations as such without being, or ever having been, in a position to make a relevant placing of some feature which, in terms of the explanatory schema I have defended, is immediately or ultimately relevant to the explanation of the type of instance concerned. To deny this would be absurd. It would be to deny, for example, that when we talk about remoter historical characters, we are really talking about individuals. But the view I am defending does not require such a denial.
For this view seeks merely to explain the notion of an individual instance of a general thing in terms, ultimately, of feature-placing. It does not at all imply that we cannot make use of this notion in situations other than those in terms of which it is explained. In fact, of course, the expansiveness of our talk about individuals is in marked contrast with the restrictedness of our contacts with them. Both the possibility of, and the incentives to, this expansiveness have an empirical ground ; in the variousness of individuals, the non-repetitiveness of situations.
But this fact may nevertheless mislead us, may make the theoretical problem of individuation look more difficult than it is by distracting our attention from an essential element in the notion of an individual instance. The problem would scarcely seem difficult for the case of an imagined universe in which all that happened was the repetition of a single note, varying, perhaps, in volume. Individual instances could then be described only as, say, " the third before now" or " the next one to come". But in such a universe the incentives to forming the notion of an individual instance would be small. We might say that, in general, what is essential to the notion of an individual instance is not what is interesting about individuals.
To conclude. I remarked earlier that the explanatory schema I have sketched points the way to a possible general distinction between the two major categories we are concerned with. To recall, first, some vague, figurative and unsatisfactory terms I have already used : the schema suggests that the notion of a particular individual always includes, directly or indirectly, that of placing, whereas the notion of a general thing does not. Now placing is characteristically effected by the use of expressions the reference of which is in part determined by the context of their use and not by their meaning, if any, alone.
And this suggests the possibility of formulating a general distinction in a more satisfactory way. We may say : it is a necessary condition for a thing's being a general thing that it can be referred to by a singular substantival expression, a unique reference for which is determined solely by the meaning of the words making up that expression ; and it is a necessary condition of a thing's being a particular thing that it cannot be referred to by a singular substantival expression, a unique reference for which is determined solely by the meaning of the words making up that expression. This specification of mutually exclusive necessary conditions could be made to yield definitions by stipulating that the conditions were not only necessary, but also sufficient. But there is point in refraining from doing so. For as we consider substantival expressions increasingly remote from the simplest cases, there may be increasing reluctance to apply the distinction at all. Nor is this reluctance quite irrational ; for the simplest cases are those which form the basis of the general distinction. (Hence, roughly, the association of particularity with concreteness).
We may admit that the traditional distinction was vague as well as unclear, and respect its well-founded vagueness in this way.
To elucidate this quasi-definition of particular and general, I add some miscellaneous comments of varying degrees of importance.
It might be objected to the conditions given that expressions like " The third tallest man who ever lived or „ lives or will live " answer to the specifications for a generalthing designation. If they did, it would perhaps not be difficult to legislate them out, by suitable amendments of those specifications. But in fact they do not. For their meaning does not suffice to determine for them a unique object of reference. It is, if true, contingently true that there is a single thing answering to such a description. This case, however, does raise a problem about how the words " expression a unique reference for which is determined solely by the meaning " are to be construed. If we construe them as " expression the existence of just one object of reference for which is guaranteed by the meaning," we may find outselves in (possibly circumventable) trouble over, e.g., " phlogiston " and " the unicorn." Yet this is the construction at first suggested by the present case.12 It will This difficulty was pointed out to me by Mr. H. P. Grice. 2G
be better, therefore, to construe them as follows : " expression the (or a) meaning of which is such that it is both logically impossible for it to refer to more than one thing (in that meaning)18 and logically impossible for the expression to fail to have reference because of the existence of competing candidates for the title". And the sense of " competing candidates" can be explained as follows : x, y and z are competing candidates (and the only competing candidates) for the title D if, if any two of them had not existed, D would apply to the third. (2) It may seem, perhaps, a more troublesome fact that the names we commonly employ for certain typess like Beethoven's Fifth Symphony, do not answer to the specifications given for a general-thing designation, although we may be more than half inclined to count such types as general things ; for these names include, as a part of themselves' or of their explanation, proper names like " Beethoven."
We have, however, an easy remedy here. We can regard the pattern of sounds in question as a general thing for which there might (perhaps does ) exist a general description the meaning of which uniquely determines its reference ; and then it will appear as the contingent truth it is that Beethoven stands to the general thing so designated in a certain special relation. This does not commit us to saying that it is a contingent truth that Beethoven's Fifth Symphony was composed by Beethoven ; but the necessity here is simply a consequence of the fact that we ordinarily and naturally refer to the general thing in question by means of an expression which incorporates a reference to a particular individual who stands in a special relation to it. Analogous considerations apply to many other types. Of course, the alternative is always open to us of declining to apply the criterion in such cases.
It is clear that numbers, if we apply our criterion to them, will emerge as general things. But this is a result 18 This qualification allows for the possible case where there is no con- venient unambiguous designation of the general thing in question ; but is not strictly necessary since an unambiguous designation could always be framed.
which will disturb few, and will certainly disturb no one Who continues to feel the charm of the class-of-classes analysis.
If we choose to apply the test to facts, we get the not wholly unappealing result that, e.g., the facts that 2 + 2 = 4, that all crows are black and that crows exist (in one use of " exist ") are general things, while the facts that Brutus killed Caesar and that all the people in this room are philosophers are particular things. For propositions, of course, the result is similar.
The distinction will correspond roughly to the old distinction between those propositions (or facts) which are " truly universal" and those which are not. In the case of facts and propositions, however, we may well feel a very strong reluctance to classify in this way at all ; and, if we do, there is no reason why we should struggle to overcome it.1
Some points of more general significance remain.
As historical evidence for the general correctness of this doctrine, we may note that Russell who, for so large a part of his philosophical life, showed an anxiety to equate meaning and reference in the case of names, finally inclined to the conclusion that the only true names are those of universals.
We do not, of course, need to adopt his idiosyncratic use of the word " name," in acknowledging the correctness of his implied view of universals.
It will be clear that the quasi-definition I am suggesting has points of contact with some of those more familiar ways of marking the distinction which turn out to be more or less unsatisfactory. For instance, it will not do to say that general things do not have spatio-temporal positions and limits, whereas particular things do.
Some general things, those of appropriate categories, like gold, 14 What I have said here of facts and propositions must not lead us to suppose that we should obtain a similar result for sentences.
These, and expres- sion-*}^ generally, will emerge as general things (e.g., in virtue of the conventions for the use of inverted commas, the expression " the word ' and ' " may be said to determine, by meaning alone, a unique object of reference).
See Enquiry into Meaning and Truth and Human Knowledge : Its Scope and Limits.
do have spatial distribution ; and some may have temporal limits. It is rather that when we refer to general things, we abstract from their actual distribution and limits, if they have any, as we cannot do when we refer to particulars.
Hence, with general things, meaning suffices to determine reference. And with this is connected the tendency, on the whole dominant, to ascribe superior reality to particular things.
Meaning is not enough, in their case, to determine the reference of their designations ; the extra, contextual element is essential.
They are, in a quite precise sense, less abstract ; and we are, on the whole, so constituted as to count the less abstract as the more real.
Finally, we may, if we choose, revert to the original philosophical way of marking the distinction in terms of the concept of an instance, and give it a sense in terms of the final definition.
Instantiability, in the philosophers' sense, ends precisely at the point at which contextual dependence of referring expressions begins, or where referring expressions, as being proper names of individuals, have meaning only in a sense in which it is altogether divorced from reference.
So general things may have instances, while particular things may not.
When it has been said that a particular must be an instance of something general, and that there must be criteria of distinctness and (where applicable) of identity for individual instances of a general thing, something of central importance still remains unsaid. In giving the relevant criteria'—or sets of criteria—for individual instances of a certain general thing, we do not indicate how such particulars are brought into our discourse. Nor do we bring a particular into our discourse by mentioning these criteria. (To mention them is still to talk in general.)
We bring a particular into our discourse only when we determine, select, a point of application for such criteria, only when we mention, refer to, something to which these criteria are to be applied ; and no theory of particulars can be adequate which does not take account of the means by which we determine such a point of application as a point of application for these criteria.
In the rest of this paper I shall try to do two things.
First, I shall try to show how, in the case of certain kinds of particulars (particular instances of certain kinds of general things), the notion of a particular may be seen as something logically complex in relation to other notions (a kind of compound of these notions). That is, I shall try to produce a partial explanation (analysis) of the notion of an individual instance, for certain cases ; and then I shall try. to show how this notion, as explained for these cases, may be used in the explanation of the notion of individual instances of other sorts of general things, and in the explanation of the notions of those other types of general things themselves. So in this part of the paper (sections 6-9), no general account is offered of the distinction we are concerned with. The procedure is essentially one of indicating, step by step, how certain types of notion can be seen as depending upon others ; and it makes no claim at all to completeness. Second (section 12), this procedure is found to suggest a possible general account of the distinction we are concerned with ; though the acceptability or otherwise of this general account seems to be independent of that of the step-by-step schema of explanation. Now it might seem that the difficulty of finding an explanation of the notion of an individual instance arises from the fact that the category distinction between general and individual is so fundamental that there is nothing logically simpler, or more fundamental, in terms of which this notion could be explained.
But I think this view can be challenged for a certain range of important cases, which can then perhaps serve as the basis for the explanation of others. To challenge it successfully, we have to envisage the possibility of making statements which (a) do not make use of the notion of individual instances, and (b) do not presuppose the existence of statements which do make use of this notion. The second condition may be held to rule out general statements ; for though many general statements make no direct mention of individuals, they have often and plausibly been held in some sense to presuppose the existence of statements which do.
So what we have to consider is the possibility of singular statements which make no mention of (i.e., contain no names for, or other expressions definitely or indefinitely referring to) individual instances of general things.
Now there certainly does exist, in ordinary use, a range of empirical singular statements answering to this description.
I suggest, as examples, the following :—
It is (has been) raining
Music can be heard in the distance
Snow is falling
There is gold here
There is water here.
All these sentences contain either the material-name of a general thing (" music," " snow ") or a corresponding verb ; but none contains any expression which can be construed as serving to make a definite or indefinite mention of individual instances of those general things (i.e., falls or drops of rain, pieces of gold, pools of water and so on).
Of course, when these sentences are used, the combination of the circumstances of their use with the tense of the verb and the demonstrative adverbs, if any, which they contain, provides an indication of the incidence of the general thing in question. Such an indication must be provided somehow, if empirical singular statements are to be made at all.
But it is important that it can be provided by means of utterance-centred indications which do not include noun expressions referring definitely or indefinitely to individual instances.
Such sentences as these do not bring particulars into our discourse.
Languages imagined on the model of such sentences are sometimes called " property-location " languages.
But I think the word " property " is objectionable here because
a) the general things which figure in my examples are not properties, and
b) the idea of a property belongs, with the idea of an individual instance itself, to a level of logical complexity we are trying to get below.
So I propose to substitute the less philosophically committed word " feature " ; and to speak of feature-placing sentences.
Though feature-placing sentences do not introduce particulars into our discourse, they provide the materials for this introduction.
Suppose we compare a feature-placing sentence (" There is snow here ") with a phrase (" This (patch of) snow ") in the use of which an individual instance of the feature is mentioned. It seems possible, in this case, to regard the notion of the individual instance as something logically complex in relation to the two simpler notions of the feature and of placing.
"The logical complexity may be brought out in the following way. In making the featureplacing statement, we utter a completed sentence without mentioning individuals. If we merely mention the individual without going on to say anything about it, we fail to utter a completed sentence ; yet what the feature-placing sentence does explicitly is, in a sense, implicit in this mere mention.
So, as the basic step in an explanatory schema, we may regard the notion of a particular instance of certain sorts of general things as a kind of logical compound of the simpler notions of a feature and of placing. But what about the criteria of distinctness and identity which were said in general to be necessary to the notion of an individual instance of a general thing ? The basis for the criteria of distinctness can already be introduced at the feature-placing level, without yet introducing particulars. For where we can say " There is snow here " or " There is gold here," we can also, perhaps, more exactly, though not more correctly, say " There is snow (gold) here—and here— and here." And when we can say " It snowed to-day," we can also, perhaps, more exactly, but not more correctly, say " It snowed twice to-day."
The considerations which determine multiplicity of placing become, when we introduce particulars, the criteria for distinguishing this patch of snow from that, or the first fall of snow from the second. Of criteria of identity I shall say more in general later. It might be objected that it is absurd to speak of an imagined transition from feature-placing sentences to substantival expressions definitely designating particular instances of features as the introduction of particulars ; that it is absurd to represent this imagined transition as part of a possible analysis of the notion of a particular instance, even for these simple cases of material-names which seem the most favourable ; and that at most what is achieved is the indication of a possible way of looking at certain designations of certain particulars.
For are not the particulars as much a relevant part of the situation in which a featureplacing sentence is employed as they are of a situation in which a substantival particular-designation is employed ? To this I would reply by asking what philosophical question there would be about particulars if we did not designate them, could not make lists of them, did not predicate qualities of them and so on. What we have to explain is a certain mode of speech.
When we turn from material-names to substancenames, the attempt to provide an analogous explanation of the notion of an individual instance seems much harder. But though it is harder, it is perhaps worth making ; for if it succeeds, we may find we have then an adequate basis for the explanation of the notion of an individual instance in other cases, and for the explanation of further kinds of general things. In order for the attempt to succeed, we must be able to envisage a situation in which, instead of operating with the notion of an individual instance of a cat or an apple, we operate with the notions of a corresponding feature and of placing.
Ordinary language does not seem to provide us, in these cases, with feature-placing sentences.
And it might be argued that the idea of such sentences was, in these cases, absurd.
For (1) it might be pointed out that an all-important difference between such things as snow and such things as cats lay in the fact that different instances of snow are, in a sense, indefinitely additive, can be counted together as one instance of snow ; while this is not true in the case of instances of cats ; and it might be suggested that herein lay a reason for the possibility of feature-placing sentences in the case of snow and for their impossibility in the case of cats.
And (2) it might be added that we have no name for a general thing which could count as the required feature in the case of cats. It is true that we speak of the cat in general ; but " the cat " ranks as a species-name, and the notion of a species as surely presupposes the notion of individual members as the notion of a property involves that of individual things to which the property belongs or might belong. It is also true that we may speak of an instance (specimen) of a cat, as we may speak of an instance of gold ; but here what follows the phrase " an instance of" is not, as " gold " is, a generalthing name which could figure in a merely feature-placing sentence, but an expression which also serves as an indefinite designation of an individual.
Does not all this strongly suggest that there could be no concept of the " cat-feature " such as would be required for the analysis to work, that any general idea of cat must be the idea of a cat, i.e., must involve criteria of identity 'ind distinctness for cats as individuals and hence the notion of an individual instance ? These objections have great force and importance ; but I do not think them decisive. For they do not show that it is logically absurd to suppose that we might recognise the presence of cat or signs of the past or future presence of cat, without ever having occasion to distinguish one cat from another as the cat on the left, or identify a cat as ours or as Felix.
The second argument merely reminds us that the resources of our language are such that on any actual occasion of this kind we in fact use, not a partitive noun, but the indefinite forms (" cats " or " a cat") of the articulative noun. But this fact can be explained in a way consistent with the advocated analysis (see section 10). Nevertheless, these arguments show something. The point about the species-name, for example, is sound ; the notion of a species, like that of a property, belongs to a level of logical complexity we are trying to get below. Second, and more immediately important, the first argument shows that if there is to be a general concept of the cat-feature, corresponding in the required way to the notion of an individual instance, it must already include in itself the basis for the criteria of distinctness which we apply to individual cats. (Roughly, the idea of cat, unlike that of snow, 9 Cf. Price, Thinking and Experience, pp. 40-41, on identity of individuals and of characteristics.
would include the idea of a characteristic shape). But to concede this is not to concede the impossibility of the analysis. It is worth adding that sometimes we do find verbal indications of our use of feature-concepts such as those we are trying to envisage ; as, e.g., when we speak of " smelling cat " or " hunting lion," using the noun in the singular without the article. There might seem to exist a more general objection to this whole procedure.
For it seems that it would always be possible in practice to paraphrase a given feature-placing sentence in use, by means of a sentence incorporating indefinite designations of particular instances ; e.g., " There is gold here " by " There is a quantity of gold here " ; " Snow has fallen twice " by " There have been two falls of snow " ; " There is snow here—and here " by " There are patches (expanses) of snow here and here " ; and so on. And if sentences incorporating definite or indefinite designations of particular instances bring particulars into our discourse ; and if statements made by the use of featureplacing sentences are equivalent to statements made by the use of sentences incorporating indefinite designations of particular instances ; then do not feature-placing sentences themselves bring particulars into our discourse ? But this argument can be turned in favour of the explanation it is directed against.
Suppose there is a statement S made by means of a feature-placing sentence ; and an equivalent statement S' made by means of a sentence incorporating an indefinite particular-designation ; and a statement T made by means of a sentence incorporating a definite designation of the particular indefinitely designated in S'. Now only if a language admits of statements like T can it admit of statements correctly described as I have described S'. (There are no indefinite designations of particulars where there are no definite designations of particulars.) But a language might admit of statements like S without admitting of statements like T. So the existence of statements like S', in a language which admits of both statements like S and statements like T, is not destructive of the analysis, but is a proof of its correctness.
If the argument so far is acceptable, then at least in the case of some materials and some substances, we can regard the notion of an individual instance as partially explained in terms of the logical composition of the two notions of a feature and of placing.
When we turn to properties and qualities, we may make use of a different kind of explanation which is also, in a sense, the completion of the first kind. I shall not, that is to say, try to explain the notion of individual instances of anger or wisdom or red in terms of the logical composition of a feature, such as anger or red, and placing. But nor shall I maintain that it would be wrong or impossible to do so. We might think of such general things as anger (or red) not primarily as qualities, properties, states or conditions of persons or things, but primarily as instantiated in, say, situations (or patches) which acquired their status as individuals from just such a logical composition.
But though this is how we might think, it is not, for the most part, how we do think. It is natural, rather, to regard those general things which are properly called qualities, conditions, etc., as belonging at least to the same level of logical complexity as the idea of individual instances of the kinds we have so far been concerned with ; to regard them, that is, as feature-/?^ things, the incidence of which, however, is primarily indicated, not by placing, but by their ascription to individual instances of material or substantial features the incidence of which is primarily indicated by placing.10 We have seen that the notion of an individual instance of some materials and substances can be regarded as a logical compound of the notions of a feature and of placing. We have now to see the ascription of a quality (etc.) to such an individual as an operation analogous to the placing of a feature. Indeed, we may find in the possibility of this operation the point—or one important point—of that logical composition which yields us the particular. The individual instance of the simply placeable feature emerges as a possible locationw These remarks, of course, apply only to some of the things correctly called properties, states, qualities, etc. 2
point for general things other than the feature of which it is primarily an instance, and hence as also an individual instance of these general things, its properties or qualities or states. One might exaggeratedly say : the point of having the idea of individual instances of material or substantial features is that they may be represented as individual instances of property-like features. The individuals are distinguished as individuals in order to be contrasted and compared. Other notions call for other treatment. I consider two more. (a)
I mentioned, at 4 (iii), a rather special class of individual instances of properties or property-like things. We form the notion of such an instance when, for example, we speak not of a man or an action as an instance of wisdom or anger, but of the wisdom of Socrates as an individual (a case of wisdom) or of Jones' present mental state as an individual instance of anger.
Here the notion of the individual instance can be seen as a new kind of logical compound, namely, a compound which includes as elements both the notion of the general thing (property) in question and that of the material or substantial individual which is an instance of it ; it may sometimes include a further element of temporal placing (cf. " his present state of anger "). (b) Instances of events, processes and changes I have so far scarcely mentioned.
Most of our most familiar words for happenings strike us essentially as names for the actions and undergoings of individual instances of material or substantial features. But there is a difference between these happening words and quality or state-words. A wise man is an instance of wisdom, but a dead or dying man is not an instance of death. Only a death is that. As regards such happening-words as these, then, we have to see the idea of an individual instance as reached by a kind of logical composition analogous to that considered in the paragraph immediately above : an individual instance of a material or substantial feature is an element in the compound.
But these, though perhaps the most important, are not the only kinds of happening-words.
9. The general form of these explanations may be roughly indicated as follows. The notion of placing a feature is taken as basic, as consisting of the logically simplest elements with which we are to operate. It is pointed out that neither of these elements involves the notion of an individual instance, nor therefore the notions of certain types of general things, such as properties and species ; and it is shown that the idea of operating solely with these simplest elements can be made intelligible for certain cases. (Features in fact of course belong to the class of general things ; but so long as we remain at the feature-placing level, they cannot be assigned to it ; for there is nothing to contrast the general with.) From this basis we proceed by composition and analogy.
The designations of individual instances of (some) material and substantial features are first introduced, as expressions, not themselves complete sentences, which include placingindications ; and, complementarily, certain types of general things (e.g., properties and types of happening) are introduced as items the designations of which do not include placing-indications and which are ascribed to material or substantial individuals. The ascription of such a thing as a property to a substantial individual is represented simply as an operation analogous to the placing of a feature ; so no circularity attends the word " ascription." Individuals of certain other types {e.g., events happening to substances, states of substances and " particularised" qualities) are then introduced as the designata of expressions which include the designations of individuals of earlier types, and hence indirectly include the notion of placing.
There are many types of individual and of general thing besides these here considered. Some may admit of analogous treatment ; and it might be possible to introduce others, on the basis already provided, by other methods of construction and explanation. But every introduction of a particular, in terms of such a schema, will either directly contain the notion of placing or will preserve, by way of individuals already introduced, the original link with this notion.
Of course the value of this suggestion, as it stands, is small. For the notion of an individual instance extends itself indefinitely, by way of far more complicated connexions than I have so far indicated ; and the limits of plausibility for the kinds of construction-procedure I have used would, no doubt, soon be reached, if they are not already overpassed. Nevertheless, I think this sketch of a procedure has certain merits :
Some of the difficulties which attend any attempt to elucidate the category-distinction between the particular and the general arise from the fact that these two classes include so many different category-distinctions within themselves.
This fact creates a dilemma for the theorist of the distinction. On the one hand, he is tempted, in a way illustrated at the beginning of this paper, into drawing distinctions which indeed separate one or more sub-categories of one class from one or more sub-categories of the other, but which fail to yield the desired result if applied over the whole field.
Or, on the other hand, in the effort to escape from this domination by irrelevant category differences, he is tempted by the prospect of a purely formal distinction, drawing for this purpose on the terms and concepts of grammar or of formal logic.
But distinctions so drawn can only seem to succeed by forfeiting their formal character and silently incorporating the problematic category-distinction.
The present procedure offers at least a hope of escape from this difficulty. For it fully allows for the differences between types of general thing and of individual ; and instead of producing one single explanation, the same for every case, it offers a serial method of explaining later types of general or particular things on the basis of earlier ones, while preserving a continuous general differentiation between the two major categories in the course of the explanation.
Too much must not be claimed for the suggested procedure, however • in particular, it must not be thought that it has been so described as to provide a criterion for the distinction we are concerned with.
Another characteristic of the schema of explanation is that it accords a central place to the notion of an individual instance of certain kinds of general things, viz., of material and substantial features. This (see section 8) is not an essential characteristic ; it could be modified. But there is reason to think that it corresponds to our actual way of thinking ; that these individuals are the " basic particulars." Why this should be so, and whether it might not be otherwise, are questions which I shall not now consider.
Finally, while not itself providing a criterion of general and particular, the schema points the way to a possible general distinction which might be defensible even if the procedure which suggests it should prove unsatisfactory. This general distinction I shall outline in section 12. Before I do so, some further points remain to be considered.
Something further must first be said on the subject of criteria of identity for individual instances of a general thing. We saw (5 (2)) how in many cases the question of the criteria of distinctness and identity of an individual instance of a general thing was incompletely determined when the general thing was named. This "was particularly evident in«the case of some properties and was evident also in the case of materials. Where substance-names were concerned, however, this indeterminateness seemed not to exist ; when the name was given, the criteria were fixed.
And this was connected with the fact that in these cases there seemed to exist no true general-thing name, apart from expressions which ranked as species-names and obviously presupposed certain definite criteria of identity for individual members. As far as criteria of distinctness are concerned, this raises no particular difficulty.
We saw, for example, how the idea of a simply placeable feature might include—might indeed be—the idea of a characteristic shape, and in this way provide a basis for criteria of distinctness for individual instances of the feature. But it is not so easy to account for the apparent determinateness of criteria of identity. The explanatory schema advanced required that we should theoretically be able to form concepts of some substance-features which were logically prior to, and independent of, the corresponding concepts of an individual instance of such features ; and this requirement seems to clash with the apparent determinateness of the criteria of identity for such individuals. A parallel answer to that given in the case of criteria of distinctness is theoretically available, but is unattractively unplausible.
If we reject this answer, and cannot find an alternative, then we must at least radically revise, though in a not unfamiliar direction, the basis of the explanatory schema. (The difficulty is essentially a more specific form of that encountered already in section 7). I think, however, that an acceptable alternative can be found. For in all cases where a feature-concept can be assumed to be possible, the criteria of identity (and of distinctness) for an instance of the general thing in question •—or the sets of such criteria, where there is more than one set—can be seen as determined by a combination of factors, viz., the nature of the feature itself, the ways in which the feature empirically manifests itself in the world, and—to adopt a possibly misleading mode of expression—the kind of incentives11 that exist for having a notion of an individual instance of the feature in question.
The rele\4ance of this third factor even, perhaps, gives us the right to say that there is something arbitrary about the criteria we adopt, something which, given the other two factors, is—in at any rate a stretched sense—a matter of choice. In extreme cases this is obvious. Even those who had witnessed the whole of the affair under discussion might, for example, give varying answers to such a question as : Is this the same quarrel going on now as was going on when I left ? The answer we choose may depend on just what distinctions we are interested in ; and one can imagine many situations 11
What I mean by " incentives " here may be illustrated from the con- venience of the institution of property. Suppose there is a general feature,
for this example, and many different things which might influence us. There may, on the other hand, be very many cases of features where the adoption of a certain particular set of criteria of identity (and distinctness) for their instances is so utterly natural that it would seem to be stretching the phrase " matter of choice " intolerably to apply it to them.
But, even in these cases, the naturalness may still be seen as depending on the combination of factors I mentioned ; and, if we bear this in mind, we can sometimes imagine the possibility of alternatives. (Here is a question which might with advantage be explored for many different types of case.) It seems reasonable to view substantial features as cases of this kind. If this view, is acceptable, we can find in it an explanation of that difference between substance-names and certain other true general-thing names to which I have several times referred.
Given a true general-thing name, like " gold " or " wisdom," the question of the criteria of identity of its instances cannot be answered until the kind of instance is specified, by such a phrase as " apiece of gold " or " a wise action".
But where one set of criteria of identity is peculiarly dominant, its adoption peculiarly compelling, we find no such non-committal general name in current, adult, unsophisticated use.
All that we might wish to do with it, we can equally well do without it, by the use of the indefinite singular or plural forms of the ordinary substancename (e.g., " a horse " or " horses "). 11.
It is, perhaps, necessary to guard briefly against a misunderstanding. Of course, I am not denying that we can very well use individual-designations as such without being, or ever having been, in a position to make a relevant placing of some feature which, in terms of the explanatory schema I have defended, is immediately or ultimately relevant to the explanation of the type of instance concerned. To deny this would be absurd. It would be to deny, for example, that when we talk about remoter historical characters, we are really talking about individuals. But the view I am defending does not require such a denial.
For this view seeks merely to explain the notion of an individual instance of a general thing in terms, ultimately, of feature-placing. It does not at all imply that we cannot make use of this notion in situations other than those in terms of which it is explained. In fact, of course, the expansiveness of our talk about individuals is in marked contrast with the restrictedness of our contacts with them. Both the possibility of, and the incentives to, this expansiveness have an empirical ground ; in the variousness of individuals, the non-repetitiveness of situations.
But this fact may nevertheless mislead us, may make the theoretical problem of individuation look more difficult than it is by distracting our attention from an essential element in the notion of an individual instance. The problem would scarcely seem difficult for the case of an imagined universe in which all that happened was the repetition of a single note, varying, perhaps, in volume. Individual instances could then be described only as, say, " the third before now" or " the next one to come". But in such a universe the incentives to forming the notion of an individual instance would be small. We might say that, in general, what is essential to the notion of an individual instance is not what is interesting about individuals.
To conclude. I remarked earlier that the explanatory schema I have sketched points the way to a possible general distinction between the two major categories we are concerned with. To recall, first, some vague, figurative and unsatisfactory terms I have already used : the schema suggests that the notion of a particular individual always includes, directly or indirectly, that of placing, whereas the notion of a general thing does not. Now placing is characteristically effected by the use of expressions the reference of which is in part determined by the context of their use and not by their meaning, if any, alone.
And this suggests the possibility of formulating a general distinction in a more satisfactory way. We may say : it is a necessary condition for a thing's being a general thing that it can be referred to by a singular substantival expression, a unique reference for which is determined solely by the meaning of the words making up that expression ; and it is a necessary condition of a thing's being a particular thing that it cannot be referred to by a singular substantival expression, a unique reference for which is determined solely by the meaning of the words making up that expression. This specification of mutually exclusive necessary conditions could be made to yield definitions by stipulating that the conditions were not only necessary, but also sufficient. But there is point in refraining from doing so. For as we consider substantival expressions increasingly remote from the simplest cases, there may be increasing reluctance to apply the distinction at all. Nor is this reluctance quite irrational ; for the simplest cases are those which form the basis of the general distinction. (Hence, roughly, the association of particularity with concreteness).
We may admit that the traditional distinction was vague as well as unclear, and respect its well-founded vagueness in this way.
To elucidate this quasi-definition of particular and general, I add some miscellaneous comments of varying degrees of importance.
It might be objected to the conditions given that expressions like " The third tallest man who ever lived or „ lives or will live " answer to the specifications for a generalthing designation. If they did, it would perhaps not be difficult to legislate them out, by suitable amendments of those specifications. But in fact they do not. For their meaning does not suffice to determine for them a unique object of reference. It is, if true, contingently true that there is a single thing answering to such a description. This case, however, does raise a problem about how the words " expression a unique reference for which is determined solely by the meaning " are to be construed. If we construe them as " expression the existence of just one object of reference for which is guaranteed by the meaning," we may find outselves in (possibly circumventable) trouble over, e.g., " phlogiston " and " the unicorn." Yet this is the construction at first suggested by the present case.12 It will This difficulty was pointed out to me by Mr. H. P. Grice. 2G
be better, therefore, to construe them as follows : " expression the (or a) meaning of which is such that it is both logically impossible for it to refer to more than one thing (in that meaning)18 and logically impossible for the expression to fail to have reference because of the existence of competing candidates for the title". And the sense of " competing candidates" can be explained as follows : x, y and z are competing candidates (and the only competing candidates) for the title D if, if any two of them had not existed, D would apply to the third. (2) It may seem, perhaps, a more troublesome fact that the names we commonly employ for certain typess like Beethoven's Fifth Symphony, do not answer to the specifications given for a general-thing designation, although we may be more than half inclined to count such types as general things ; for these names include, as a part of themselves' or of their explanation, proper names like " Beethoven."
We have, however, an easy remedy here. We can regard the pattern of sounds in question as a general thing for which there might (perhaps does ) exist a general description the meaning of which uniquely determines its reference ; and then it will appear as the contingent truth it is that Beethoven stands to the general thing so designated in a certain special relation. This does not commit us to saying that it is a contingent truth that Beethoven's Fifth Symphony was composed by Beethoven ; but the necessity here is simply a consequence of the fact that we ordinarily and naturally refer to the general thing in question by means of an expression which incorporates a reference to a particular individual who stands in a special relation to it. Analogous considerations apply to many other types. Of course, the alternative is always open to us of declining to apply the criterion in such cases.
It is clear that numbers, if we apply our criterion to them, will emerge as general things. But this is a result 18 This qualification allows for the possible case where there is no con- venient unambiguous designation of the general thing in question ; but is not strictly necessary since an unambiguous designation could always be framed.
which will disturb few, and will certainly disturb no one Who continues to feel the charm of the class-of-classes analysis.
If we choose to apply the test to facts, we get the not wholly unappealing result that, e.g., the facts that 2 + 2 = 4, that all crows are black and that crows exist (in one use of " exist ") are general things, while the facts that Brutus killed Caesar and that all the people in this room are philosophers are particular things. For propositions, of course, the result is similar.
The distinction will correspond roughly to the old distinction between those propositions (or facts) which are " truly universal" and those which are not. In the case of facts and propositions, however, we may well feel a very strong reluctance to classify in this way at all ; and, if we do, there is no reason why we should struggle to overcome it.1
Some points of more general significance remain.
As historical evidence for the general correctness of this doctrine, we may note that Russell who, for so large a part of his philosophical life, showed an anxiety to equate meaning and reference in the case of names, finally inclined to the conclusion that the only true names are those of universals.
We do not, of course, need to adopt his idiosyncratic use of the word " name," in acknowledging the correctness of his implied view of universals.
It will be clear that the quasi-definition I am suggesting has points of contact with some of those more familiar ways of marking the distinction which turn out to be more or less unsatisfactory. For instance, it will not do to say that general things do not have spatio-temporal positions and limits, whereas particular things do.
Some general things, those of appropriate categories, like gold, 14 What I have said here of facts and propositions must not lead us to suppose that we should obtain a similar result for sentences.
These, and expres- sion-*}^ generally, will emerge as general things (e.g., in virtue of the conventions for the use of inverted commas, the expression " the word ' and ' " may be said to determine, by meaning alone, a unique object of reference).
See Enquiry into Meaning and Truth and Human Knowledge : Its Scope and Limits.
do have spatial distribution ; and some may have temporal limits. It is rather that when we refer to general things, we abstract from their actual distribution and limits, if they have any, as we cannot do when we refer to particulars.
Hence, with general things, meaning suffices to determine reference. And with this is connected the tendency, on the whole dominant, to ascribe superior reality to particular things.
Meaning is not enough, in their case, to determine the reference of their designations ; the extra, contextual element is essential.
They are, in a quite precise sense, less abstract ; and we are, on the whole, so constituted as to count the less abstract as the more real.
Finally, we may, if we choose, revert to the original philosophical way of marking the distinction in terms of the concept of an instance, and give it a sense in terms of the final definition.
Instantiability, in the philosophers' sense, ends precisely at the point at which contextual dependence of referring expressions begins, or where referring expressions, as being proper names of individuals, have meaning only in a sense in which it is altogether divorced from reference.
So general things may have instances, while particular things may not.
Posts
No comments:
Post a Comment