I think Jim grasps one of the most important things (to me at least) about my approach to Frege in the book. At least part of the picture I wanted emerges if we think of Frege, at least for an important part of what he gave us, as engaged in a research program. Any research program carries assumptions of some sort or another. Insofar as they are assumptions, they are not immune to collapsing, taking the program with them. But Frege’s assumptions, or the ones I have in mind here, are anchored in grammar, in that sense of ‘grammar’ in which (later) Wittgenstein insists it is precisely, and at least in greater part, what philosophy is about.
The aim of the program I have in mind is to locate, as he puts it, the (most general) laws of being true. So truth (and falsehood) are to be the central quantities of concern to the theory aimed at; the central ones which the laws sought govern. There is a certain phenomenon which is the home of being true. I now call it (though not in the book) nonfactive representing. What such representing represents, in the central case at least, is the way things are, or their being. It represents this as being a certain way there is for it to be (or not). What makes it non-factive is that here there is not a requirement on representing things as thus and so that they so be. (Compare factive meaning, e.g., that the lights are turned off may mean that the restaurant is closed.) And it is just this non-factivity which brings with it (here) the packet truth and falsehood. For it is just this which leaves open that pair of possibilities, representing things as they are, and as they are not.
Following further the model of a scientific theory—say, Newtonian mechanics—the parameters to whose values the laws we seek are sensitive come in rerum natura surrounded by ‘noise’, as the workings of force and mass come in a cloud of factors from which Newton’s laws prescind, for example, air resistance—something which misled Aristotle (for a start). Frege’s research program builds in two contributions to screening off noise. They enter with the idea that he is interested only in the most general laws of being true (vide e.g., 1983/1897: 137) He thus insists that the phenomenon which interests him is both thinker- and topic-neutral.
To see what thinker-neutrality does for Frege we need some points of grammar. The first thing to note is that a verb such as ‘represent’ (at least in represent-as) comes in aspects, as many verbs do. Consider the verb ‘to open’ as in ‘He opened the door’. Also as in ‘This key opens that door.’ The man here is credited with a historical performance, involving, inter alia, some key turning (or substitute therefor). The key itself hardly puts on performances (unless, perhaps, rusting, or getting jammed in the lock is a performance). But it does open the door: that is, such is what it is for, and such is what will happen if used correctly and all is in order (e.g., no chewing gum in the lock). I will call such use of the verb ‘inert aspect’. The contrasting aspect, as in ‘She opened the door’ I will call ‘engaged’.
So far, non-factive representing is represented in the inert aspect. It need not be. There is also an engaged aspect. Thinkers can engage in non-factive representing. They can do so in either of two ways: first, auto-representing (taking-that, fearing-that, assuming-that, etc.). Second, (just for contrast) allo-representing (thought-expression). But recall Frege’s project of screening out noise. Since (one half of) the goal is thinker-neutrality, plausibly part of the noise to be screened off (for the sake of the most general laws of being true) is a thinker’s contribution to the representing (in particular in thought-expression). So it is the inert aspect of the verb which is of interest here. From which arises what Frege labels ‘Gedanken’.
‘Gedanken’ and not ‘Denken’. It is worth reflecting on the ambiguity of the English nominalization of ‘thought’. In fact, it has two sources (thus is really a pair of nominals). One is thinking, either an activity we are sometimes said to engage in, or simply a condition, e.g., the thinking of ‘think-that’. Anyway, the business of thinkers. The other is the thinkable (as a body or mass), for example, thought on the topic of ‘green’ energy, or trout fishing in America—or, just thinkable in general. Such is what Frege selects out from this assortment in his notion Gedanke—a countable, specifically a thinkable—thus, with a bit of care, a truth-or-falsehood (a bit of non-factive representing). I will return presently to that ‘bit of care’.
So a Gedanke is to be a countable bit of non-factive representing, truly or falsely, inert aspect; in other terms a thinkable. One thing Frege is concerned about making part of the programme is that a thinkable enjoys a certain ontological priority.
I do not think that concept formation can precede judgment because this presupposes an independent existence of concepts, whereas I think concepts arise through the decomposition of judgeable content. (1882, letter to Marty).
He also holds (vide 1919: 274–5) that when one decomposes a Gedanke there must be at least one proper predicative part (thus predicable). A predicable predicates in general of n items, the value of n depending on the predicable. In which case a Gedanke might be thought of as a predicable of n-ity null.
So far we are only halfway through the programme. We have covered thinker neutrality. There remains topic neutrality. Here we start from the idea that a thought has decompositions: proper parts, each doing a determinate part of the work done by the whole thought (each engaged in contributing to non-factive representing, all jointly (in the context of the decomposition) doing precisely that representing which the whole does. The point would be proper thought parts (e.g., a predicable which predicates of a thing that it is thus and so) are recurrable, a given one recurring in an indefinitely extendible range of thinkables, two occurrences, each in a different thought, making for a respect in which the two are the same, such sameness being something into which a law of logic might in principle be able to sink its teeth.
So laws of being true may be thought of as governing relations between (sets of) thought-decompositions. However, they are not allowed to appeal to any topic-specific contents of those thought parts. Very well. Redact (black out) the contents, leaving behind only the type of proper part (e.g., predicative, denotative, etc.) and the structure of the whole decomposition. Laws which govern such structures can count as topic-neutral, thus as logic’s business. With which we have identified the parameters, or variables, to which a law of logic may be sensitive, and thus logical forms (redacted decompositions) as the stuff of which logic treats. And, in first prescinding from thinkers, and then abstracting to forms, we have shunned the noise (all equivalents of air resistance in mechanics) which obscures the lawfulness of laws in the business of being true. I turn now to three questions Jim poses.
1. Why Cannot a Thinkable be blue? The first step by way of response is to introduce the notion of thinkable. This is another series of steps:
1. A thinkable is a (in fact principal) participant in what I call non-factive representing. By ‘non-factive’ I mean: a notion represented on which it is not required for so doing that things are represented. Benno’s notoriety may represent years of hard work. But only if there were such years. Sid’s shiner may represent a collision with a door (or its after-effects). But only if there was such a collision. Pia may represent Benno as notorious just as well even if he is not. Thus in this form of representing, it is possible to represent things as they are not, but also as they are. Call what the first false is, what the second true.
2. A thinkable represents a certain aspect, or understanding, of the verb which I call inert. The clue lies in the last 4 letters. Representing in this sense is something there is to do, a way of thinking available to be engaged in. This aspect of the verb (as I am calling it) is far from peculiar to represent. It is the aspect in which a key opens a lock (even while incapable of auto-locomotion). If you want to open that lock, this is the key for you. Similarly for a stain remover and removing stains.
While we are at it, a brief taxonomy of non-factive representing. One division into types lies along the inert and the ‘engaged’, the doing of what the inert makes available for doing. Inert representing, for example, commits to nothing. Engaged representing might do. One might simply wonder whether Sid loves lager. But one may also be certain that he does.
Engaged representation comes in two main forms. In one of these, thought-expression represents authoring. On the other, it is for one’s thinking to take on a certain shape, to be sensitive in some determinate way to things being (or not) a certain way there is for things to be. Here ‘things’ operates as a mass term, not a plural. It is short for things being as they are.
3. A thinkable brings truth into question. It is that by which truth is brought into question at all. That is, it is precisely what raises/fixes a determinate yes-no question, ‘True?’, where without it there would be no such question (at all), or at least none would as yet have been raised. Here Jim and I may differ on a hermeneutic matter. Here is the untranslated text:
Ohne damit eine Definition geben zu wollen, nenne ich Gedanken etwas, bei dem überhaupt Wahrheit in Fragen kommen kan. (Frege, Reference Frege1918: 61)
Taking thinkable to be near enough Gedanke in Frege’s sense (about which more to come), I translated this: a Gedanke (aka thinkable) is just that by which truth can come into question at all (that is by which there is an answerable question, ‘True?’). At the least, Jim has qualms about the ‘just’. Its trigger in the text is, of course, the ‘Überhaupt’, which I translate ‘at all’. So a Gedanke is that by which there is a question of truth at all: any less than what the Gedanke supplies and there is no such question.
I have understood this as follows: Suppose that for a Gedanke to be the one it is, inter alia, for it to have a given feature. Then without that feature, it would not pose the question of truth it does, and would not make the same demand on being true. Conversely, suppose that without a given feature it would still pose the question of truth it does. Having that feature is not part of its being the Gedanke that it is. Thus, too, thinkable.
So there is a certain minimalism about thinkables. They are in one business, representing truly or falsely, a business in which, if true they are the truth in question. (The truth is that Sid loves lager.) Mutatis mutandis falsehood. They are just what they need to be to play the role in the business they are assigned.
4. Such minimalism shows up in the fact that the predicable true is an identity under predication: Predicate it of a thinkable and you get that thinkable back. The truth (or not) of the thinkable Sid drinks is settled by settling whether Sid drinks. So the truth (or not) of the thinkable It is true that Sid drinks is also settled by whether Sid drinks. If he does, then the thinkable he does is true, so the thinkable that that thinkable is true is also true. And the thinkable is just that over which true is an identity under predication. For example, sentences are obviously not this. (The sentence ‘Sid drinks’ and the sentence ‘It is true that Sid drinks’ are different sentences.
5. A corollary: A thinkable represents things as a way there is for things to be. For it to be that thinkable is just for it to represent things in that way. So there are two different thinkables only where there is a difference in the way each represents things as being, thus in what is demanded by each of the ways things are.
So far, I have identified a phenomenon, non-factive representing, and then introduced a notion, thinkable, which I claim is one of a pair of central protagonists in the business of such representing (of representing truly or falsely), it in the role of representer, its companion represented being the way things are (or things so being). I am now arguing of that notion, thinkable, that thinkables are ineligible for being colored, or conversely, what is coloured is ineligible for being a thinkable. Admittedly the notion of thinkable is introduced here by stipulation. As Frege stresses (vide 1893: xiii–xiv), stipulation cannot bring anything into existence. It can only zero in on something there is any way to think about. This means that stipulation can never be proof absolute against collapse. As objects of sight can never be proof absolute against ringers. Here, though, I can see no determinate form of collapse in need of dismantling. If Jim differs, I await his case.
There are, in fact, worries concerning Frege’s Reference Frege1918 introduction of Gedanke. Namely, he does this in terms of another proprietary notion, Sinn. A Sinn is a feature of a bit of language, e.g., a ‘concept word’ (to use his vocabulary). The Sinn a Gedanke is meant to be would be a feature of a sentence. A sentence is a dedicated means of thought-expression, the role of its Sinn would be, inter alia, in determining which thought the sentence expressed when. But a sentence’s means of making such recognizable are not all available as parts of a thinkable. For example, a thinkable cannot make use of deixis. So the Sinn of a sentence, if it is going to be a Gedanke, will need to prescind from at least some of what is to be understood as to how the sentence speaks of what it does. And it remains to say how such prescinding is to be done. Perhaps Sinn is a misbegotten notion. I take it, though, that for the present purpose such matters can be bracketed. Caveat emptor.
Caveat given, I return to present business. I do not think that Jim’s worries are with my stipulations per se. Or if they are, I await hearing them. As I understand him, his brief is that thinkables should have properties non-intrinsically and that he cannot see why one of these should not be blue. I agree with Jim on the first count. Of course, thinkables have properties other than those which make them the ones they are, or those they have intrinsically. Stronger, they had better do if they are going to be objects of thought themselves (as they have been so far). I will try here for an overview of these. But no, a thinkable could not be blue. Being a thinkable excludes being blue, and being blue excludes being a thinkable. One may approach the issue from either side.
As to the non-contingent properties of thinkables, I suggest that they come in three main sorts.
First, thinkables relate to each other in such relations as, centrally, entailment. Speaking with Frege, for example, some may be Merkmale of others: That something is a blue silk ribbon may entail that it is silk. (Vide Frege, Reference Frege1919: 114) Some may depend on further facts. That Sid was in Rome last Saturday may entail that his chair at Stammtisch was empty that evening.
For the second I invoke again my taxonomy of non-factive representing. A thinkable represents in inert aspect. But the inert is, in one way, parasitic on the engaged. A key opens a lock (is so usable) if so using it is, at least in some weak sense, in the cards. There must at least be such a thing as what using it would be like. Mutatis mutatis for the thinkable. A thinkable may represent things as such that Benno smokes Murads. But if so then such is something one may commit to, e.g., in being convinced that, Benno being the kind of poseur he is, such is what he would do. So thinkables (inert representers) bear relations to engaged representers. Sid fully expects the smell of Murads he will in fact encounter at Benno’s soirée. Inert representers, anyway thinkables are, by grammar, made precisely to fit as objects of such verbs as ‘suspect’ ‘doubt’ ‘fear’ ‘take it’, and may stand in such relations to given thinkers, populations, etc.
Thinkables also bear relations to allorepresenting. Centrally, a thinkable may be the thought expressed in a given episode of thought expression. It may be just what Benno whispered in Pia’s ear. In Fregean terms, it is precisely designed to relate to whatever there is for an Absicht to attach to that the item, or its production, be so to be understood. Scarf-tied Dutch style may mean that the drinks are on Benno tonight.
Inert non-factive representing can be thought of as an abstraction from its engaged relatives. For it to fit in its appointed places in these engagements it is eo ipso to be receptive (in context) to certain ranges of properties which it would have (if at all) non-intrinsically (On that occasion at Benno’s, e.g., Sid could have kept it to himself.)
So far there is this much in common to the properties a thinkable may have non-intrinsically. The above may not be exhaustive. But, on the evidence, such does not admit a thinkable such properties as being blue. (Nor will it be any help at all to decide that a certain thinkable may be written only in blue ink.)
Whatever you can mention, there are myriad ways for it to be or not contingently (so myriad such ones it is). But which ways-to-be these may be is decided by what is intrinsic to the item.
It is intrinsic to being coloured that it takes up space, or anyway spacetime. (Also it interacts causally with the environment.) Representing something as some way for it to be (inert aspect) does not take up spacetime. It has no such dimensions. (For something to be square is not itself square.) Nor will bleach fade it. So a thinkable cannot be blue. Nor can any predicable (inert aspect). Color is simply not a variable by which predicable predicates might vary. There is nothing that would count as it so being.
It is intrinsic to a thinkable (or any predicable (way for something to be)) that it generalizes—admits of indefinitely many different cases of a relevant something so being (casu quo not). For something to be blue is one way for something to be, on a par with any other. But what is blue, a sample of being as predicated does not predicate, is not itself a way for something to be.
Thinkables, as Frege stressed, are not visible, audible, or tangible. That they are not seen in one way in the fact that there is no such thing as how Zorn’s lemma is spelt. A difference in how two items sounded would not make a difference in how each represented things as being (if either did so at all). Were predicables to sound or look like something, differences in how a predicable sounded or looked would settle nothing as to how it represented its predicants as being.
Note: In Jim’s contribution (p.??) he proposes something I might have in mind as to what bars thinkables from being coloured. Part of the diagnosis commits me to this: “[A thought’s]only non-identifying properties are those any thought would have in having its identifying ones.” There is an ambiguity here in the ‘its’. On one reading, if a thought were blue, then any thought would be blue (whatever its ‘identifying properties’). On the other, any thought which had the aforementioned identifying properties would also have to be blue (in both cases, of course, if any thinkable could be coloured at all). I hope I made clear in the above text why neither of these plays a role in the case just made.
2. Why Cannot A Thinkable Bear Content? I start here with words of Frege’s at the outset of “Der Gedanke”:
So is a picture, purely as a visible, tangible thing, really true? And a stone, a leaf is not true? One would not call the picture true if no intention is attached to it. The picture is meant to represent something. Nor would a Vorstellung be called true in itself, but only with reference to an intention, that it should correspond with something. (1918: 59)
Pia tells Sid: If you find a green leaf on your pillow in the morning, that means that I’m off to Faro for the weekend. If it is a brown leaf, that means, ‘Not au revoir, but adieu.’ A leaf is a leaf is a leaf. But sometimes it is to be understood as representing things as being thus and so. Leaves can be bent for such tasks. What the ‘thus and so’ is—which way for things to be— is left entirely open. Suppose that, in place of the leaf, Pia had just woken Sid up and uttered these words: ‘I’m taking the Porsche to the shop.’ This, too (the performance), might be meant, by prior arrangement, to be understood to represent things as being in whatever way, e.g., such that Benno is coming for dinner, or the plants need watering, or that she will be lecturing in Munich next week, order alte Schnorrer has run out of lager, or ‘not au revoir, but adieu’.
Thus it is wherever there is something to which an ‘Absicht’ is attachable. But where, or to what, is an Absicht attachable? What I claim, and claim on Frege’s behalf, is that such content-bearing belongs in a particular corner of the general phenomenon of non-factive representing, namely that variety of engaged representing, thought expression. And since this occurs in an environment, coming as it does in historical (at least loosely datable clockable) episodes, in point of grammar an Absicht-bearer (that to which it attaches) must be capable of (recognizable, perceptual, ‘tangible’) presence or absence in given environments on given historical occasions.
One might cite various reasons for this. But the main one, I think, is this: For thought expression to occur, the thought thus expressed must be recognizable (in principle at least) as the thought it is. Otherwise, we have a failed case, one of mere would-be thought-expression. The role of an Absicht-bearer in the present case is to aid recognition. To do which, the idea is, it itself must be recognizable—as present or absent on an occasion—so that (by prior arrangement, or convention, or whatever) its presence can serve as a tipoff to the way things are to be understood to have been represented in the relevant performance of attempted thought-expression (as also, if needed, to the fact that such an act of representing is occurring, or to be understood to have occurred) at all.
That leaf, like that lantern in the church steeple, does not itself engage in non-factive representing (but only in representing of the factive sort—it may thus mean, e.g., that autumn is upon us). The leaf is not per se to be understood as bearing any message. But it can be recruited to do so. Well-formed expressions of a language may make themselves easily recruitable by, and for, speakers of that language, but then for a bespoke assignment, as, in point of English, ‘Sid smokes’ is for use in calling a contextually determinate someone a smoker, whatever would then be meant by that. Thus a caveat in advance. If the bespoke use for a well-formed expression is narrow enough, one might say that the expression itself bears an Absicht in the relevant sense, even though strictly speaking it is only instances (e.g., inscriptions) of the sentence that are liable to be present or absent on occasions).
Part of the idea so far then is that considering the task to be performed, an Absicht bearer must itself be recognized (or recognizable) as present (or absent selon le cas) if its bearing an understanding is to be of any relevant use, i.e., any use in making recognizable how a given performance represented things as being. And I take it that it is just this idea at which Jim balks. He suggests that inert non-factive representing, too, can bear an Absicht in the present sense. In which he crosses a border from engaged representing, in particular, in the present case, what requires an author (a thinking being) to inert representing, which admits of none. And thus from the spatiotemporal locatable (at least, in the case of books, the author’s exertions) to what eo ipso, has no spatiotemporal career.
A thinkable which represents Sid as a smoker is one item among an inexhaustible collection of peers. But it is not anymore, casu quo less, present on any historical occasion than it is on any other. So in no way can its presence be deployed by an author as a tipoff to what the author is up to. In any sense in which we might speak of it as having a presence, its presence on one occasion does not distinguish that occasion from its absence on any other. It is present or absent tout court.
So Jim has his work cut out for him to say the least. In a footnote to his comments, he produces what I take to be an example proving the justice of what he wants to say. This may well show what it is he has in mind. But if so, then the example’s failure is transmitted accordingly. And his example does fall short of the goal. Here are its working parts. First (as we all know) the Illuminati are spread worldwide (everywhere to be found hiding under rocks). Some are deaf, some blind, some perhaps both. None are polyglottish enough for there to be a common language (though presumably each ‘speaks’ at least one). So communication between them, or between them and others, will have to be bespoken. Sometimes an audience must be spoken to, sometimes written to, sometimes have a message tapped out in Morse code on the audience’s wrist, etc. Thus an aspirant communicator with one of these would have to assign values to two parameters: first, relevant handicaps if any (deaf, blind, etc.). Second, relevant language. Moreover, a language being systematic in the way it is, and the secrecy illuminati-communication demands requires that changes be worked by fiat on what some sentences of the relevant language would normally express. E.g., a sentence which in its language speaks of lawns as green, maybe, by fiat, to be taken within the projected communication as speaking (positively) of the fitness of the Illuminati for inheriting the world.
Right away we can see how this misses the mark. The relevant point is just that none of this gets us out of that small corner of non-factive representing, to wit, thought expression. And, complications notwithstanding, we all knew all along that this corner is entirely open to working the relevant changes. By stipulation ‘Bububu’ can be, in the mouth (or fingertips) of an illuminated expresser bent to the task of expressing the thought that Illuminati rock. But such does not extend the range of that susceptible to Absicht-attachment beyond what we already knew it to be.
Jim and I disagree on a thesis which, concisely stated, could be put thus: “Non-factive representing (predicables, n ≥ 0) are contents to be born, (hence possible objects of Absichten in Frege’s sense), but cannot function as bearers of content. But now expansion of the notion bearer may be called for. Suppose that ‘to bear content’ simply meant: to be to be understood as representing things (or a thing) as being thus and so. To illustrate, take the thought (i.e., thinkable) that Sid smokes (is a smoker), assuming for the moment that this is either true or false. It is to be understood as representing things as being thus and so (a certain way).
But to understand it as so representing things and to understand it (full stop) are one. Understanding it just is understanding what way it represents things as being. To understand it otherwise (if such were possible at all) would be to misunderstand it—at best, take it for some thinkable it is not. It cannot be so to be understood. Punkt. Perhaps there are episodes of someone grasping it, but there are no episodes of someone (correctly) taking it to bear some other understanding (that is attached to it by an Absicht) on some occurrence of it. (Moreover, an occurrence of it could only be in the context of some compound thought, or some set of thoughts, e.g., connected to some other set in a way which makes a transition from first to last truth-preserving. An occurrence of a thinkable cannot have a spatiotemporal profile.
A caveat. Imagine this case: Sid has long smoked a pack a day (or more), and still does. But each cigarette with a gun at his head, each puff with revulsion. (The beagle’s revenge.) Is Sid a smoker? Perhaps to the oncologist’s eye, yes. Oncology takes little interest in mens rea. But to society’s eye perhaps not. Exactly part of the horror of the whole thing, from which we may wish to liberate him by force.
So there is an oncological understanding of being a smoker, and a ‘social’ understanding of which mens rea matters. We might also find, on reflection, that nothing in our understanding of being a smoker tout court chooses between these. So far, then, it looks as if we may not have a genuine thinkable truth-or-falsehood. Suppose that we now decide that the thinkable we identify in the thought of being a smoker is, say, the oncologist’s—or again, if you prefer, ‘society’s. We might think of the way for something to be which we now speak of in speaking of being a smoker as an enrichment of the understanding we had before of what we then spoke of. Which might be put: In bearing the understanding being a smoker now does, it still bears the understanding what once went by that name did, but then elaborate in re what would count, what not as so being. If we did so speak, then in representing Sid as we now would in (say) falsely representing him to be a smoker we also represent him as being as he neither would or would not be on that old understanding, only this last mere proper part of this first.
Such might be one way of describing such things. I do not plan here to treat the general phenomenon just illustrated. Anyway, it provides some sense for some notion of bearing understandings on which a thinkable may bear one understanding in bearing another. The point of raising the issue here is to bring out how different this would be from what Jim has in mind in trying to bring Absicht-attachment across the border from engaged representing (specifically that of thought-expression) to the realm of the inert. For one thing, the present proposal does not yield any denizen of the realm of inert non-factives which yields for an Absicht any collection of alternative understandings of a given thinkable from which to choose.
There is thus at least this contrast between Absicht-attachment and proposals such as the present. First, a thinkable is, and remains, precisely not something admitted to being understood in any of many ways. It is (vide 1918: 60–1) what raises a determinate yes-no question, ‘True?’. Second, a thinkable is not a device for creating any representation of some further sort. Again, it is not susceptible to being present or absent in a sense thus required. It does not achieve the existence of further representing as what does admit of bearing an Absicht does. In this last case, the Absicht-bearer, in making recognizable what is going on in a case of thought-expression helps bring it about that given thought was expressed. In the case just described, the only representers ever in the picture are just potential outcomes of such maneuvers.
3. Laws and Laws: In the book, and the first part of this reply, I depicted Frege as carrying out a research programme in search of the laws of being true. What I sketched is modelled very closely on the path towards laws of mechanics, e.g., chez Newton. One must filter out ‘noise’ to arrive at parameters to which the laws of the phenomenon may be sensible. E.g., it is by filtering out noise that Newton was able to correct Aristotle’s idea that motion was a quantity, like petrol, expendable.
Laws come in many flavors. In the preface to Grundgesetze (1893), Frege points to one distinction between two sorts of law. A law of mechanics differs from a law of logic in that the first governs a natural phenomenon and the second does not. But both sorts are alike in that what they identify are facts of factive meaning (or consequence). In the Newtonian case, for example, applying a given force to a given mass has consequences for the acceleration there would then be. In Grundgesetze’s logic, or standard first-order, the truth of a thought has the consequence that anything else (materially) entails it: for all of anything else, still, it holds.
Frege also points out that for any law of what is, there is a corresponding law of what is to be, or at least one of a certain shape. If failing to brush one’s teeth twice a day has the consequence that they will fall out before one is 65, then one must brush his teeth twice a day in pain of having his teeth fall out. Similarly, Frege tells us in 1897 (1983/1897 p. 139), that a law of logic tells us something about how one must think to reach the goal truth. In present terms, one must restrict himself to transitions from thoughts to thoughts by the permissions logic grants (or by transitions underwritten by factive meaning in some other way, e.g., by dentistry or zoology).
A law of what is to be has the superficial form of a law of Sittlichkeit: Obey it or else. But that same form may be just a grammatical riff on a law of what is: ‘You must brush your teeth twice a day or they will fall out’ says no more than that, as things are, not so brushing will result in their falling out—a law of what is. And laws of logic are of this sort. Derive Q from the mere fact that P implies Q and you have wandered onto a non-certified path, liable to lead to falsehood. If you do so anyway, well, that is your business. If, however, you aim for truth then caveat emptor. It is wrong to conclude that an object is a mammal merely because all whales are. For all of which, your conclusion may be true.
So not every law of what is to be contains, like a law of Sittlichkeit, an unqualified ought or must. Some contain at most qualified ‘ought’s. You ought to conform to the laws of chemistry if you hope to get hydrogen out of water. And, ceteris paribus, you ought to conform to the laws of logic where your aim is truth. (Logic has limited governance over factive meaning.) Here, perhaps, there is a special rider to add: Complete disregard for what logic’s laws state may lose one his eligibility for participation in thought in countables altogether.
In any event, laws of motion are not made psychological laws simply because if you want to make a body accelerate you had better apply force as thus recommended. Even less (logic’s laws not being laws of nature) are logic’s laws psychological just because by a grammatical riff they are expressible in imperative form.
Now, though, Jim claims to have discovered a way in which logic may still be, to a certain extent, psychology (and that its being is nothing to scandalize Frege). I’m not sure quite what he has in mind. But it apparently has something to do with Frege’s claim that for one not to be confined to an ‘inner world’ he must win for himself (or be gifted with) an environment. As Jim seems to read this, the situation is this: There is a variety of things to think about. Some of these belong to an ‘inner world’. Some belong to an environment. What a thinker chooses to think about is thus up to him/her. S/he is free to choose from each of two menus. However, Frege tells us, that logic (or anyway philosophy) strongly recommends a certain choice: ‘Invest heavily in the environment, only lightly at best in your inner world. There will be mental health benefits of doing so.’ With this, we do see logic (or serious philosophy) meddling in the psychological. After all, what can pure logic have to tell us as to what will benefit mental health, and what not?
But I think this is to misunderstand radically what Frege has to tell us in these passages (and others) in “Der Gedanke.” The key point here is that for Frege there is no such choice to make. Thought is per se of an environment (though not entirely a spatiotemporal one). So in effect, what Frege is giving us here (notably in the later parts of “der Gedanke”) is a private language argument. I will not rehearse it in full here. But for one thing, the point comes out in a simple way in the very idea of a predicable of n-ity null (and of that as the truth-or-falsehood). At n-ity null every predicable predicates of the same thing (the way things are, aka the environment (outer world)). Just so that no further specification is needed as to of what any given zero-place predicate is predicating. No predicable could be predicated of two different inner worlds, or one such world twice—except as Frege tells us, where (if ever) predicating of an outer world is what counts as so doing.
How to read the imagery? Gaining an environment, I suggest, is gaining something to think about. It is not as though there are no non-thinkers (snails, e.g.) in what there thus is to think things of. Nor should imagery of ‘unlocking’ an environment suggest that there is some particular key, a new neural module, or something of the sort that does the work. Such a commitment would be psychology. But I do not think it is one Frege is making in the relevant passages. The point is rather that thought of an environment (aka the outer world) and thought in terms of truth-and-falsehoods (engagement in non-factive representing) are one. To be eligible for one is eo ipso to be eligible for the other. To be locked within an inner world, were this to befall one, would be not to be a thinker at all. Which is not a bit of psychology.
The question one might thus ask is what it is about non-factive representing that requires an environment (aka the outer world). Equally, what would be missing in an ‘inner world’? One thing to note here is that a predicable involves recurrence. Such belongs to its ‘reaching beyond the particular case’. (For further unfolding of ‘reaching beyond see my reply to Shieh.) In effect, a predicable assigns a particular content to the same. For A and B to be as predicated by a given predicable, ∏—for both to be the relevant way there is for something to be) is for them to be in a certain respect the same. And if A’s being as it is what would count as something being ∏, and ditto for B, then again there is something both are, and understanding on which A and B are the same. Thus the truth of any thinkable is susceptible to bearing on that of any other. Thus that truth can restore a profile in factive representing (and require one).
Lightning, it is said, strikes only once in any given place. Suppose so. And suppose further that last night lightning struck Pia’s fig tree. Then this will not happen again. Which does not rule out that tonight it will strike der alte Schnorrer’s potted palm. Suppose so. Now there is a sense in which the Kneipe’s palm and Pia’s figueira have suffered the same fate. In another direction, that very same lightning bolt might be responsible for a power outage in Pia’s street, in turn, responsible for the clock on her microwave blinking (and even for this going unnoticed into the next day).
By such means, the material is generated for confirming or disconfirming claims that Pia’s figueira was struck. And so on. Compare being confined to an inner world. Here one may still confront a passing scene, an endless episode of being presented to oneself as no other is, and as one is presented to no other. But without recurrences of the sort generated by the idea of predication, there remains no way of carving out these determinate countable ways for things to be (at least for one). Frege speaks of such objects of experience in images of private property. They have an owner, one particular experience, and are coeval with his/her experience. Here, Frege stresses, there is no material for engagement with truth or falsity, nothing the truth of a response to such experience could be hostage to. Here, to borrow a now popular cliché, all is always and only a matter of ‘It is what it is.’ Evidement!
I mean here only to sketch a picture, be it the one I claim Frege is (rightly) pushing around the end of “Der Gedanke.” But in terms of what we have so far I end with an example of Frege’s which for long I had failed to understand. It concerns a patient in pain and his doctor. The doctor is a bit uncertain as to the cause and/or treatment, so he consults another doctor. It ends thus
Both doctors have as the common object [of their thought] the pain of the patient, whose bearer they are not. From this, we can see that not only a thing but also a Vorstellung [private object] can be the common object of the thought of people who do not have this Vorstellung. (1918: 73).
Vorstellungen belongs to that constant flow of being presented to oneself as only one can be and as no one else is. We, from our third-person perspective, can cut this flow into temporal parts. If we had no outer world (environment), we could, of course, not do this. Nor could the owner of the Vorstellungen, if locked in an inner world, do this. But suppose that we and she/he do have an outer world. This does not stop us from having Vorstellungen in Frege’s ‘psychological’ sense.
On the contrary, it allows us to make something of them. Pain is a notorious case of being presented to oneself as one is presented to no one else. (Wittgenstein, at one point, considers what a counterexample to this might look like, with at best meagre results.) Two doctors and a patient—the sufferer plus two other thinkers—can have this as a common object of thought. So in that sense, it is not (or need not be) the sole property of its sufferer. What long puzzled me was why Frege thinks he can have this for free rather than as the conclusion of an argument. But the point is that it is enough for him (and us) to say: Such is just part of what pain is, or is to be understood to be. Such is a possible understanding for those (such as us), already with an environment to think about, that is, for anyone eligible to engage in nonfactive representing.
That such understanding would be unavailable to a being confined to an inner world only confirms the hegemony of the environment over the thinkable. From this perspective Frege is right to treat the first sentence of the above quote as banal, thus reinforcing the idea that there might be two different kinds of thinkable, one about an environment and another about something else. What there is to say about that passing scene available only to its owner is that insofar as there is anything to say about this, it is by drawing that scene, at least insofar as there are such things to say, into the environment.
Not that there is no flow of being presented to oneself as to no other. Of course, there is. Rather when such flow is articulated into elements, its sufferer relinquishes, eo ipso, exclusive property rights in thoughts about them.
Open access
