2. This seeming paradox, however, dissolves fairly easily when we realise just what it is that `sameness assertions' assert.

a) Consider what it is that we are saying about the state of affairs (the world out there) when we say:

i) Your car is the same as mine-

ii) The same statement can be made in many languages-

iii) He lives in the same house as I lived in when I was a child.

We are not saying that your car is my car, though we are saying that it just is that statement and we are saying that two people live (at different times) in one house.

But it should not be assumed from this that we are employing different kinds of sameness; there are no different kinds of sameness.

In each case something is the same (just that thing) and something (else) is sufficiently different to distinguish the one instance of that thing from the other instance of it.

In i above the make, colour, size, age, etc, which count as that car just are what the are - but one instance of them belongs to me, and the other instance of them belongs to you. In ii the statement just is that statement - but there are many makings of that statement and these are distinguishable from each other because different modes of stating are employed (think back to the statement/sentence relationship). In iii what is being compared is that house at one time and that same house at another time. It is claimed that the house he lives in is the house I lived in - but not that his living in that house is my living in that house.

b) The (pseudo) problem arises from a mistaken assumption that sameness is `two-termed', that for there to be a sameness X must be the same as Y. But this would amount to X being Y, which is plainly absurd (as we have seen). In fact sameness is `three-termed' (just as giving is three-termed - refer back). For there to be a sameness (an instance of sameness) X must be the same as Y in respect of Z. My car is the same as your car in respect of its shape, size, colour, etc, the stating in English is the same as the stating in German in respect of what is stated; my domicile years ago is the same as your domicile now in respect of the physical surroundings in which they occur.

In each case the Z just is what it is. X is one instance of Z-ness and Y is another instance of Z-ness and instances are necessarily particular instances of general (universal) kinds. What makes the comparison, and therefore the assertion of the sameness (the identity of the Z) possible are the non-Z properties of X and of Y respectively. The `total X complex' is clearly distinguishable from the `total Y complex' though Z-ness is common to both. All we are doing is noting that Z (that one, same Z) is instantiated in both an X context and a Y context. There is a sense in which, in all assertions of sameness, the different X contexts and Y contexts are `taken as read'.

c) People who have followed the exposition in a) and b) above should realise that the `popular' distinction between `the same one' and the `same kind', though it is often useful as a clarifying device in communication, has no philosophical validity at all and, indeed, can often be a source of confusion.

If somebody said, for instance, `My teacher always wears the same clothes', he might well be asked `Do you mean the same actual garments or simply garments which look the same?' Or consider: `My hearthbrush caught fire the other night; it had been such a good one that I went straight to the shop and bought the same one again'. Here it is quite obvious from the context that a different piece of wood is involved - yet there is, nevertheless, a continuity of that hearth brush. Plainly the `identity of' the piece of wood involved is being regarded as irrelevant to the identity of the hearth brush. But, if I said `You might not believe it but this is the same brush I had when you visited ten years ago' then, from this context, it is quite obvious that it is the piece of wood that I am talking about.

Now, because on the one hand we have solid, tangible things like shirts and hearth brushes and, on the other hand, nebulous and insubstantial things like presented appearances (`looks') and functions, it is natural enough that people think of the first lot in `same or different one' terms and of the second lot in `same or different kind' terms. But there are two good reasons why we must recognise that this is a spurious distinction -

i) Not all `objects of attention' are physical objects. Those three-dimensional `things' which occupy volumes of space are by no means the only `think-about-ables'. In the (alleged) `same one/same kind' distinction, the `same one' is always (think about it!) the same physical object. Thus we run into problems in two ways: when the appearance and function of that physical object which was once an X are so changed that it can no longer count as an X [If, by amateur carpentry I have turned my wardrobe into a bookshelf, it cannot be the same wardrobe since it is not a wardrobe at all, notwithstanding that it is the same physical object, the same pieces of wood] and when the `object' in question is not a physical object at all. We can talk about `numerical sameness' (same one) and `qualitative sameness' (same kind) in an understandable way when we are discussing peas in a pod or cars on an assembly line - but not when we are discussing dreams or stories or problems or purposes or arrangements or statements.... But in all of these cases we can intelligibly ask `Is this the same as that? - the same dream in several different dreamings, the same story in several different versions, the same problem experienced by different people or at different times....

ii) The `distinction' is misleading because it is quite unnecessary - all identifications involve both `numerical difference' (of the instances) and `qualitative sameness'- of whatever universal quality (complex) it is, that they are instances of. Consider:

a. Things don't have identities; people identify things - they `see this as' another instance of that complex.

b. Their doing this just is their particularising universals - this (in this X context) is the same Z as I encountered previously (in that Y context). And the Z in question can be any think-about-able.

Now look again at the three examples in a. above:

i) This total presented appearance is instantiated by the physical object we call your car and the same total presented appearance is (also) instantiated by the physical object we call my car.

ii) This construction of bricks, mortar, timber, etc instantiated in a 1990's, occupied by-Jones context is the same construction of bricks, mortar, timber, etc as is (also) instantiated in a 1930's, occupied-by- Brown context.

c) The crucial question is always: What is the concept which is being particularised? What is it that the X and the Y are both instances of? But, whatever the answer to this, the instances (as instances) must be numerically different from each other - how else could they be instances? - and the universal, the class of which they are instances, must be just what it is that one (same) class - how else could they both be instances of it?

d) So - every identification must be the identification (recognition) of one instance, as distinct from all other instances, of a class. So - what is identified is always, necessarily, class-membership. We gain our (universal) concepts by abstraction; we apply our (universal) concepts (in propositional beliefs) to those particulars (those objects of attention) we recognise (identify) as instances of them.

3. It should now be useful to think again about connotation and denotation - and about `thisness' -

a) A denotation is always of a connotation - and (as seen in 1. d) ii above) the connotation can be such that the denotation happens to be one (only) instance. The denotation of `Prime `Minister' must include all those individuals who ever were Prime Ministers of any countries. But the denotation of `Australian Prime Minister in 1986' happens to be Bob Hawk - and only Bob Hawk. Here note that `Australian Prime Minister in 1983' has a denotation of two - Hawk and Fraser.

It should be noted also that, although every term must have a connotation (or it could not function within language rules), that connotation can provide a denotation of nil. We can, for instance, talk intelligibly about fire-breathing, purple- spotted dragons but it seems very unlikely that there happen to be any.

It should be apparent that, when we attempt to identify (to `isolate') any particular object in a descriptive way, we do so by producing a connotation which is sufficiently extended to ensure that the object in question is the only member of its denotation.

b) But - it should also be apparent that no connotation, however extended, could necessarilly produce a denotation of one only - i.e. it is not possible to give a description which could apply only to one instance (of a class). This is why we speak of what the denotation is, or happens to be, never of what it must be.

This becomes clear when we recall that the connotation (the description) must be a `catalogue of universals' (of nesses, of kinds, of class-names). And any universal could have any number of instances (any class could have any number of members). So that, in `narrowing down' our denotation by extending our connotation, each additional ness (universal) we add to the `catalogue' includes the totality of its instances - any of which could be also instances of any or all the other nesses in that catalogue. So - since what we are seeking is not simply a description which this object meets but a description which exclusively identifies this object and no other, we overcome the problem by adding `thisness' to our catalogue of nesses (our connotation, our description). We say: not only is it an instance of the co-instantiation of all of these characteristics; it is this particular instance.

c) But - should we get away with this? Only if we are quite clear about what it is that we are getting away with! As we have seen, there are no bare `thises'; thisness is not a characteristic of objects, it is a function of our thinking about them. `This' must be (understood as) this something (i.e. this instance of X-ness). Consider:

i) What we are attempting to identify (to render unique) is one particular part of objective reality (fact) - but all that we (can) have to identify it with are class-concepts (universals). But we do not feel satisfied to identify it as a member of a class (however complex that class may be) we feel that we need to identify it as this particular member of that class.

ii) Whereas thought (about the world) is `in universals' (nesses), the world (the objective reality thought about) is `composed of' particulars - each object of it (each fact) just is what it is and nothing else. And what makes it what it is can only be the totality of its relationships with every other fact - i.e. with fact as a totality. Now, each of these `other facts' could (in principle) be thought about in universal terms as an instance of a class. So that - theoretically - the connotation which would genuinely render this instance unique (the only actual instance of that connotation) would have to include, in universal terms, the entire catalogue of this object's relationships with every other object in the universe. [Since there can be no criterion for how many different objects (facts) there are, it is doubtful if we should even say `theoretically'!] It must be remembered that everything which is true of X is true of X in exactly the same way; there is no hierarchy of `truths'. It is no more true of this pen that it is blue than that it is at present X centimetres from the South Pole. So, to put it another way: to render X unique descriptively would be to list everything which is true of X. And plainly nothing could ever count as that.

iii) So, what we are doing when we `add thisness to the connotation' (when we say `this blue pen') is making the `this' `stand for' the totality of those additional factors (unknown to us) which do in fact render this instance of blue-penness unique - i.e. that set of characteristics which conjointly make this instance of blue-penness different (discernable) from every other instance of blue-penness. [If you have followed this closely, and grasped the argument, you will see the sense in which the identified instance always becomes thereby a sub-class, one assumed to have one member only. So it remains true that all identification can only be of classes.]

4. At this stage we should recapitulate - attempt to formulate and/or illustrate a few general principles about identity that have emerged.

a) Nothing has an identity; people (and other animals) identify things - i.e. recognise them as the things which (for the idenitifiers in question) they are. And what people recognise is that a particular object is a particular instance of a particular (universal) class. For Jones to identify his own house is for Jones to see X as another instance of that customary complex which (Jones) counts as his own house. In doing so, Jones is `imposing' an identity upon a building; he is identifying it with all other (past) instances of what he sees as that complex.

b) The fact that the house has been re-roofed, re-painted, extended by a porch, does not (necessarily - or even probably) prevent Jones from seeing it as his (same) house. For him, the style of the roof, the colour of the paint, the attachment or non-attachment of a porch are not essential parts of what counts as that house. But, of course, the house is different - and, if the differences in question were seen by Jones as relevant, then, for Jones, it would be a different house. And, even if it had not been re-roofed and re-painted, there would still be some differences between the instance observed today and the instance observed yesterday - if nothing else, that one was today and the other yesterday. There must be something to `render discernable' the different instances, or they would not be instances.

From which it follows that, however unlikely it might seem, Jones could, quite rationally, see his house today as a different house from the building he called his house yesterday. There are not, and cannot be, any rules for what must count for somebody as the `essential characteristics' of `this X'. The question to ask, therefore, is not `Is this the same as that?' but `In what respects is this complex the same as that complex?' - i.e. `By what criteria could it be asserted that this and that are both instances of the same thing?'

Here consider again: X is the same as Y in respect of Z. If, for the identifier, the Z in question includes the `essential properties' for `being this thing' - then, for that identifier, X is Y. - e.g. for Jones, this house is the house he inhabited yesterday.

c) This is, unfortunately, complicated by the inevitable `M/N gaps' in our customary complexes. (Here refer back to the discussion in 2. f) of paper No. 4). Jones will accept it as another instance of his house if the Z in question includes any M of the N properties which he regards as being essential to its `being this thing'.

And, because of the `any M of N factor`, situations frequently arise in which the identifier himself is not quite sure whether to say X is the same as Y or not - i.e. he can't quite determine what he is counting as the Z in question.

Here consider what counts for you as this being the same book as that ( `Is this the book I was reading yesterday?') The answer must depend upon what you identify (conceive) as an instance of bookness - i.e. on your concept of book(ness). Someone might say `No, its not the same book (that one had marmalade spilt on page 2) but it is the same edition (so it doesn't matter)'. [Note, in passing, the relevance of what matters (to somebody for some purpose)]. Someone else might say `No, you're wrong, it isn't actually the same edition - but it is the same story'. And, to this, it could be retorted `Yes but- the same story could be used by a dozen different writers' - `No, you mean the same plot'....

d) Now, there are disagreements here and they are about whether X is (should be regarded as) the same as Y or not - but the disagreements are not about what words are printed on what pieces of paper (there most likely is complete consensus on that). So - if two people are arguing about whether X is or is not the same story as Y (when they are agreed about what words are on what pieces of paper), the basis of their argument can only be either the significance of `same' or the significance of `story'

- Examine these -

i) Is sameness necessarily always the same sameness? Can we assume that each of us has one only concept of sameness and that our concepts are all effectively equivalent? The answer to this is that we not only do assume it; in order to reason and to converse we must assume it.

Look again at what is being asked: Are all instances of sameness instances of the same sameness? Compare: Are all instances of bestiality instances of the same bestiality? The answer to this second question is `No' - on the grounds that people do use the term `bestiality' for two quite different things - cruel or brutal behaviour to other people and sexual intercourse by people with non-human animals - neither of which in any way implies the other. If the answer to the question about sameness were `No' - then this could only be on similar lines - that people use the term `same' in various ways which in no way imply each other.

But (and here think about it carefully) - if this were the case, then the question itself would be unintelligible (or at least, ambiguous, which plainly it is not) since, if there were distinctly different uses of `same', people would not know what was being asked in this case, which use of `same' applied. So we have the odd situation that it would be a question only if the answer is `Yes' - But, that being so it isn't really a question at all: it is merely a `question-type' sentential form of the assertion of a necessity. (Here refer back to the discussion of sentence/statement relationships in 4. d) of paper 6).

ii) So, since the people disagreeing are not at variance in their uses of `same', they must be at variance in their uses of `story' - have different concepts of storyness. For Brown, what counts as a story is a particular sequence of symbol-tokens (of sentences); for Jones, what counts as a story is the description of a particular series of events. And, plainly, what counts for each of them as a story (as storyness) must govern what counts for each of them as this story (this particular instance of storyness).

e) All of this simply underlines the point that objects don't have identities; people identify objects (recognise situations) according to their thinking-patterns and focal points of interest. So there is no right answer (in an objective way) to questions like `Is your woodheap the same one as you had last year?' The chances are that it is different lumps of wood but in the same place (and with the same function - it is a woodheap). If particular lumps of wood count as a woodheap, then these same particular lumps of wood count as this woodheap; if place and function are what counts as a woodheap then the same place and the same function are what count as this woodheap.

Since for most people it is the latter, there is unlikely to be a misunderstanding (here think back to communication-breakdown) of the question `Do you still have the same woodheap?'. But there are cases in which practically everybody uses a given term for two categorially different concepts. Compare:

i) I have sent my uniform to the cleaners (so I must wear mufti).

ii) The janitors at all the branches wear the same uniform (that uniform).

It is immediately apparent that, if a uniform were the kind of thing we could send to the cleaners (a particular physical object) then it would not be possible (thinkable) for several different people to be simultaneously wearing a (that) uniform.

So - before answering the question: `Is it the same uniform (as Jones wore on Saturday)?', we must ascertain whether what is being asked is `Is this another instance of that set of garments?' or `Is this another instance of that set of visual characteristics?'.

Only when this has been ascertained can we consider (on an `any M of N' basis) whether for us this does count as the same one (another instance of that complex).

f) Here again we strike the difficult point (encountered in 3. b) and c) above) that the `this' is always `this instance' - but merely being this instance of uniformness is not enough - all uniforms are instances of uniformness - what we need is a particular instance of that particular (instances of particulars can only be particular instances) uniform. Now, we cannot talk of `instances of instances' - that would be plainly nonsense. Instances are of classes. So it follows that every Instance of a class provides a sub-class of that class. The `this' (see 3. c) iii above), is `standing for' the full set of characteristics that render the instance unique;it thereby `imparts' an extended connotation (a set of further qualifications which must be met) and, thereby, creates a sub-class - in the quite straightforward way that steamships are a sub-class of ships.

All classes have and are sub-classes (nesses are `composed of' nesses). What we identify is, necessarily, an instance of a class (a sub class). Now, since `this' stands for a set of qualifying characteristics, this-pen-ness is a sub-class of pen-ness in exactly the same way as green-pen-ness is a sub-class of pen-ness. What I identify, therefore, when I recognise this pen, is an instance of `this-pen-ness'.

It follows that the identification of this pen is, as a process of thought, no different whatsoever from the identification of this object as a pen. Once this point is properly grasped, it is seen that ldentification just is classification, which is always, necessarily subclassification since we move, inevitably, in our thinking, from the vaguer to the more precise. Identification is now seen as nothing more or less than the recognition of (instances of) customary complexes - and all the mystery goes out of `identity'.



5. Now, with all of this in mind, we shall take a slightly different approach to the question of what makes an object the object that it is, what we are asserting when we assert that it is that object.

a) We speak of objects - or just things - and also of events. The object is that which we identify - and, plainly, there are situations in which our `objects of thought' are events. If, for instance, we say `His birthday party was a great success', what we are talking about is the event, his birthday party. But, in any given context, we can distinguish what we think of as `the thing' from what we think of as the event happening to it. Events happen to things and what they must be is changes (of some kind) in those things. `The tree split down the middle' described an event happening to the tree.

[We shall later examine, when we consider perception, the epistemological inter-relations between objects and events and the sense in which what we perceive is always an event. For our present purposes, however, we can accept the simplistic thing/event relationship exemplified above]

So we are distinguishing objects - what changes happen to - from events - those changes which happen to particular objects. Here it can be useful to think in `subject/predicate' terms: The subject is (seen as) an actuality the predicate poses an assertion about that subject which is a possibility. If we ask `Is this pen running out of ink?' we are assuming the existence of the pen and posing the possibility that a certain change is occurring to it. You cannot have a change without something which is changed - but the something in question can remain what it is without that change occurring.

Now, suppose the answer is `yes' - then we can, as it were, `move up a notch'; we can ask `Is this pen-which-is-running- out-of-ink being thrown away?' So that the `event' of the first question has become incorporated into the `object' of the second question. [Compare this with the point made earlier that the particular is always, in a sense, already propositional].

But none of this alters the fact that, at any given time (within any given believing), what counts for anybody as the object (the identity) is that which he sees events (changes) occurring to.

b) Let's put this very simply indeed. If we see the event as a change in the object X, then after that event it still is the object X; if we see the event as a change of the object X, then it is a change into something else, the object Y. The change, of course, just is the change that it is; the distinction between `changes in' and `changes of' exists, as it were, in thought, not in objective reality.

For instance: in Newcastle for many years there was an annual festival called Mattara. At one stage it was deliberately changed from a rather tawdry affair involving professional showmen with stalls and roundabouts to a `citizen festival' with an emphasis on community creative activities. Suppose Jones said `What we call *Mattara now is not the same festival as that which people called *Mattara in the 70's' and Brown disputes this: `I grant that it has changed, but is still the same annual festival'. Neither of them is right or wrong. For Jones the changes include factors which he saw as the `essence of' (part of what counts as) this festival (and, therefore, every instance of this-festival-ness); for Brown the changes are simply of the accompanying characteristics (the X and Y contexts - refer back) of what he sees as what counts as that festival.

We might say, then, that for each of us, within a given thought framework or focus of believing, the identified object is that which we see as constant through a process of change, that to which the changes are occurring. Where the changes are such that there is no longer `M of the N characteristics' which count for us as that thing, then we see this as a change of that thing into a different thing (X has ceased and Y has commenced). Here think again of the woodheap and note the essentially subjective nature of identification; how far is the woodheap permitted to `wander' (as woodheaps do) before it becomes (for somebody) a different woodheap?

c) Now note again the point that everything which is true of X is equally true of X - and, that everything is constantly changing because the totality is changing. It follows, then, that in any identification of that one again (of an instance of this X-ness) we must overlook, or treat as irrelevant, some changes. If we did not then all that anybody ever could identify would be this point-instant - but it would not be identification because there could be nothing to identify it with.

Here it is important to bear in mind, however, the difference between what is the case and what is believed (by somebody) to be the case. Since it is people that do the identifying, it is beliefs, not facts, which govern the identifications. That a difference exists, therefore, `enters the identification' only if it is known ( or believed) to exist. [If the belief is false a misidentification will occur. This does not imply that there are (objectively) right and wrong identifications - only that when somebody judges that X is (an instance of) that one and subsequently discovers that, by his own criteria, X is not (an instance of) that one - he has made a mistake]

To give a very simple example of this - say that for Jones `this particular teapot' is indeed this particular lump of treated clay. He well may think he is using the same teapot as he used yesterday but feel that he has made and `identity error' when his wife informs him that she broke the teapot last night and replaced it today with `another just like it'.

d) Just as everything must be (in some respect) different from everything else (identity of indiscernables), so, in some respects, everything must be the same as anything else. Here we are back to that previous `vagueness/precision' point (compare error occurring only in some context of correct judgement).

Remember that since what we identify is a class - and thereby a subclass - i.e. to identify is to sub-classify, all classification must be at some given level of precision/vagueness. From which it follows that, insofar as Jones identifies this as oak and that as poplar, the two things (for Jones) are different things - but, insofar as Jones identifies this as tree and that as tree (and nothing more) then (for Jones) these just are two instances of the same thing. Certainly one tree is here and the other tree is there - but the hereness and thereness are simply the X-factor and Y-factor which differentiate the instances. They are different instances of the same thing (of that) in exactly the same way as my pen this morning and my pen this afternoon are (for me) different instances of the one thing - my pen.

Until people can grasp this quite clearly - and take it seriously - they will inevitably be confused about identity. We must get wholly rid of such metaphysical nonsenses as `the real identity' and ` searches for identity' (!) and recognise that identifying (recognising instances of kinds as instances of those kinds - and expecting accordingly) is what we are doing constantly throughout our conscious lives. To believe (that X is Y) is to identify.



6. Because it has been a source of considerable confusion, it is as well, before leaving our first analysis of identity, to look at the concept of resemblance. The common (and generally understood) assertion that `It isn't the same - but it's very like it' has much the same (!) status as `being almost sure' or `not quite possible'. At a `language rules' level we understand well enough - yet, if we take seriously and literally the elements of the assertion, we find category-incompatibilities. Indeed, our very understanding of (appropriate response to) such expressions is itself a source of confusion; it leads us, unthinkingly, to assume that there can be degrees of certainness, non-possible approximations to possibility - or a relationship between situations which can be characterised as likeness (or resemblance) and which is quite distinct from, and independent of, identity (sameness, one-ness).

a) Insofar as any such characteristic existed it would have to be quite different from sameness, since plainly there can be no degrees of sameness and, equally plainly, there are (as we use the term) degrees of likeness. [Here compare: there can be no degrees of possibility (nor yet of actuality) but there must be degrees of probability].

b) But it should be seen that, much as the degree of probability (according to somebody's belief) arises from the frequency (in that believers experience) of certain possibilities being actualities - so the degree of likeness (according to somebody's belief) depends upon the proportion of characteristics (perceived by him to be) shared by the two situations. In the terms we have been using: by how much of the X and of the Y is included in the Z (of which they are both instances).

Looked at in these terms, we see immediately that, when we assert that X is the same as Y, we cannot be asserting that X is Y (or we would be denying the (non Z) X-factors and (non-Z) Y factors which make the comparison possible - so we can only be saying that X is like Y. This does not introduce likenesses as some new and different characteristic; it merely describes the relationship between two complexes which are the same in respect of (i.e. are both instances of) certain characteristics. Thus, unless we asserted the identity of a certain part of X with a certain part of Y, there could be no grounds whatsoever for asserting that X resembles (is like) Y.

c) It seems likely that the interest some philosophers have shown in `resemblance', and attempts to elevate it to a queer kind of natural phenomenon, arise from a combination of thinking of identity as something that objects have and, at the same time, realising the essentially subjective nature of identifying. Since resemblances are what people observe, it has been suggested that the observing of resemblances is our way of `approaching' those identities. If we replace the (mysterious) notion of `real identities' with the more straightforward acceptance of fact being simply how it is, then this is a reasonable enough account of our `formulation of our customary complexes' (our nesses). But the observing of resemblances is then seen simply as the identification (seeing as instances of the same - of that) of more or less elements of different complexes.

It makes good enough sense to say that X resembles Y in respect of Z - so long as we imply, thereby that the Z in question just is whatever it is in both instantiations. But if we attempt to go further and assert that the Z-in-the-X only resembles the Z-in-the-Y, then we are faced with the question: In what respect, then, do these two instances of Z-ness resemble each other?

And it should be clear that, as soon as we make this move, we have embarked upon an infinite regress. Sooner or later, we must say: This just is that one characteristic, instantiated both here and there. So we might as well say it sooner!

d) Suppose it be protested: we are not talking of resemblances as some kind of `queer entities' in the, world, there to be observed; we are talking only of those situations where people are reminded of one situation by another. Here we could ask two questions: `What kind of reminding are we talking about?` and `Why does one situation remind people of another with such uniformity that they are able to develop public languages and communicate?'.

i) Reminding of the kind we often call `association of ideas' (the kind of thing psychologists are fond of) has nothing (obvious) to do with identity - or resemblance. [Here think again of `mere co-incidence'; Jones might be reminded of Brown by a thunderstorm simply because he met Brown during a thunderstorm].

ii) So the only reminding we are talking of is the recurrence of some presented phenomenon in different complex situations. And we must take `recurrence' seriously - this element of X is (for us) that element of Y. And if this were not so within the state of the common objective reality with which both Jones and Brown are appraised, then there would be no reason at all for both Jones and Brown to be reminded of the same things by the same things.

e) But here, once again, it is necessary to stress the point that we do the identifying and we must do it at some level of precision/vagueness. We do not need to know everything about X (whatever that may be!) in order to know something about X. So, when we observe that X resembles Y - that X is the same as Y in respect of Z - the Z just is what it is at the level of precision/vagueness at which we are currently operating.

It should be seen that there is a clear correlation between the degree of precision of the observation and the degree of resemblance we observe.

f) Let us, then, allow that, whereas it is perfectly intelligible to talk of two things only resembling each other, in so doing we are talking only of

i) which characteristics of each are the same and which are not - and

ii) at what level of precision/vagueness we could identify one with the other, as compared with the level of precision at which we are making our observations of the two things in question [Note that (and how and why) the oak resembles the poplar and, for most people, resembles it more closely than it resembles the pine].

It should now be clear that any notion of resemblance (or likeness) as a state of affairs distinct from and independent of identity is both incoherent and totally unnecessary.

7. Finally - to summarise some of the major points -

a) Identity is a logical notion - X implies Y and Y implies X; it is not a `feature of reality' - no aspect of reality can be another aspect of reality.

We must identify in order to think (propositionally) about reality.

b) What we identify X as can only be an instance of a class (a universal, a ness). But in identifying any instance of a class, because it is that instance, we, thereby, create both a particular (instance) and a new sub-class. We can, therefore, identify with greater and greater precision - but never with `ultimate precision'.

c) This-X-ness is a subclass of X-ness - and, as such, must have its own instances - e.g. Jones-on-Saturday-ness is a sub-class of Jones-ness. An instance of Jones-on-Saturday-ness occurs every Saturday (so long as Jones lives) - and Jones-on- Saturday-morning-ness and Jones-with-a-headache-on-Saturday-ness are subclasses of Jones-on-Saturday-ness - etc.

d) What we call resemblance is `comparative partial identity'. We will assert that X is Y or that X is like Y according to the level of precision at which we are identifying - i.e. what differences we see fit to `discount' in our identification.