In this post I want to tentatively examine particular issues in Dummett's philosophical works that might connect to similar issues in tarot history. I have spent a few weeks working on this post, conjointly with reading and re-reading those of Dummett's philosophical writings that seem to pertain. Dummett is in the tradition of Frege and Wittgenstein, those two giants of analytic philosophy in the periods 1850-1900 and 1900-1950 respectively. Fortunately, a long time ago I did study these philosophers, Wittgenstein rather intensively. But I am rather rusty, and I don't doubt that I have committed mistakes and confusions.
Here I am developing further some thoughts I initiated on a different thread, "Dummett and tarrocchi appropriati", in particular my post at
viewtopic.php?f=11&t=1008#p15043. However I am putting these further thoughts on the "Dummett and methodology" thread because they pertain to his methodology generally (as does, in part, the post just linked to) without addressing
tarocchi appropriati in particular.
GAMES AND MEANING AS USE
In Dummett's philosophical writings, I have found only one passage which actually alludes to the game of tarot to make a philosophical point. In
Thought and Reality, originally given as the Gifford Lectures at the University of St. Andrews in 1996, he says (p. 55):
A statement of the rules of a card game may begin with a specification, which may be simple and may be complicated, of the order in which the cards rank; some may be trumps and they will rank in a particular order, while the cards of the plain suits may rank in a different order. All this, however, is only a preliminary to the rules of play that follow. It is significant only in so far as the ranking of the cards figures in those rules.--for example, in determining who has won a given trick. When you know only the order in which the cards rank, you as yet know nothing about how to play the game. Similarly, when you know only the truth-conditions of sentences of a language, you as yet know nothing about how to speak the language.
The philosophical point is that an account of linguistic meaning must contain an account of the different uses to which words and sentences are put: to make assertions, ask questions, give commands, make requests, give advice, express a wish, etc. (I get this list, his most comprehensive, from his 1976 "Frege and Wittgenstein," on p. 247 of his book
Frege and other philosophers.)
The point about tarot is one that he makes independently of philosophy in
Game of Tarot p. 166:
To understand the purpose for which the Tarot pack was invented we have therefore to ask for what reason an ordered sequence of cards, of different length and composition from the ordinary suits, was added to the regular pack (…) Obviously to find an answer, we have to look at the role that these cards play in the game, on the reasonable assumption that the essential features of the game, in the various forms in which it was later played, belonged to it from the start.
I have lifted this last quote from p. 8 of Carlo Penco's essay, "Dummett and the Game of Tarot" (
https://www.academia.edu/2442994/Dummet ... e_of_Tarot), where Penco draws the connection to philosophy, in particular to the later Wittgenstein's "meaning is use" slogan. To understand the meaning of a word, sentence, or card, look at its use in the context of the human activity involving it. Penco has drawn the connections and appropriate conclusions in his article (section III, pp. 7-10). The main problem, as Penco points out, is that just as sentences can be put to different uses, so can decks of cards. In fact, games themselves have a variety of uses: for some it is entertainment, for others a diversion from worries, for others the hope of achieving some local prestige, for others a way to make money. It is only to their use in a game that occult meanings and divinatory uses are irrelevant. If it can be accepted that the lack of direct evidence of these uses implies that there were no other uses, then Dummett is right to exclude such meanings. That, of course, is the issue.
Games in the Middle Ages and later sometimes had didactic uses. For example, the game known as "the philosopher's game" (
http://en.wikipedia.org/wiki/Rithmomachy) was designed specifically to teach Boethian number theory, especially in its application to music. Chess, similarly, was thought to develop a sense of military strategy on the battlefield that would be useful in later life. That tarot also had a didactic purpose at the beginning is suggested by the fact that the trump suit had no numbers on it, requiring the players to memorize a specific sequence of familiar symbols. Dummett's
Game of Tarot argument that the designers "didn't think" to include such numbers is rather weak, given that the practice of not supplying numbers was repeated throughout the 15th century, even in the woodblock "Cary Sheet"; and in the non-standard Sola-Busca, c. 1491, such numbering was used. The practice of numbering items in a sequence was commonplace, as for example items in an account book and pages in a book.
Also, when we look at the earliest known deck, the Cary-Yale, a didactic purpose would seem to be implied by the majority of the extant cards. They reflect the subjects of Petrarch's
Trionfi and the seven virtues. Given that Petrarch's poem is an argument for Christianity and its virtues, there is a unifying principle for the cards. Only the Empress, Emperor, and the so-called "World" card do not obviously fit this principle, although they can be made to with some effort.
In this deck another purpose is evident from the numerous heraldics embedded in the imagery, namely, to commemorate the Visconti family, perhaps including its feminine partners in other families (Savoy.and Sforza at least, although the evidence is not unambiguous). Such heraldics and feminine marrige partners were indicated in the illuminated manuscript tradition of the Visconti. The Emperor and Empress might then suggest the ducal status of that family, given by the Holy Roman Empire, whatever other purpose they might have had. The "World" card similarly might have a purpose in relation to the Visconti.
Whether other imagery in other decks had other purposes, say, to allude to a non-orthodox-Christian conception of salvation or of fortune-telling (e.g. astrology). is then a matter of discussion, not to be ruled out by the fact that the primary purpose of the cards was their use in a game. However the mere presence of astrological or "occult" (i.e. known to an elite only) images in the added cards does not imply a corresponding use, as images could have been borrowed by the artist from anywhere. It is only if the borrowed images form part of a systematic interpretation of the sequence in terms of the relevant "occult" system that we can entertain such speculations.
When Wittgenstein introduced his "meaning is use" slogan, his main point was to emphasize the different uses to which language is put. He even thought of giving his book
Philosophical Investigations the motto, from
King Lear, "I'll teach you differences". And in that book, section 23, he says:
But how many kinds of sentence are there? Say assertion, question, and command?--There are countless kinds: countless different kinds of use of what we call "symbols", "words", "sentences". And this multiplicity is not something fixed, given once for all; but new types of language, new language-games, as we may say, come into existence, and others become obsolegte and get forgotten. (We can get a rough picture of this from the changes in mathematics.)
Here the term "language-game" is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life.
Review the multiplicity of language-games in the following examples:
Giving orders, and obeying them--
Describing the appearance of an object, or giving its measurements--
Constructing an object from a description (a drawing)--
Reporting an event--
Speculating about an event--
Forming and testing a hypothesis--
Presenting the results of an experiment in tables and diagrams--
Making up a story, and reading it--
Play-acting--
Singing catches--
Guessing riddles--
Making a joke, telling it--
Solving a problem in practical arithmetic--
Translating from one language into another--
Asking, thanking, cursing, greeting, praying.
It is interesting to compare the multiplicity of the tools in language and of the ways they are used, the multiplicity of kinds of word and sentence, with what logicians have said about the structure of language. (Including the author of the Tractatus Logico-Philosophicus.)
The author of the
Tractatus is Wittgenstein himself, in his earlier years.
Later in the book (sections 304, 317, 363) Wittgenstein makes similar observations, there applied to ascriptions of pain, implying that such statements are not in the same kind of language-game as ascriptions of physical states. For example, philosophical puzzles develop around statements like "I am in pain", after which Wittgenstein says:
The paradox disappears only if we make a radical break with the idea that language always functions in only one way, always serves the same purpose to convey thoughts--which may be about houses, pains, good and evil, or anything else you please.
For his part, Dummett does not deny that language can have more than one use. As I said earlier, he mentions making assertions, asking questions, giving commands, making requests, giving advice, expressing a wish, etc. ( p. 247 of
Frege and other philosophers); he only disagrees that the uses are "countless". Also, he focuses his energy on one just use, which he calls "conveying information". From that emphasis, I conclude that he thinks that such use is more fundamental than others. He thinks that the later Wittgenstein disagreed on that point (p. 247f of above essay).
Fortunately, we do not have to enter that discussion, since it seems obvious, given the facts that Dummett adduces in
Game of Tarot that objects (i.e. cards) added to a set of objects already used to play games would also be used to play games. It is also obvious that this does not exclude other uses: if nothing else, entertainment, or prestige, if one is a good player. So surely this game-playing use is key. This thesis, and the approximate earliest known date, had already been enunciated by A.E. Waite (
http://www.sacred-texts.com/tarot/pkt/pkt0104.htm; see especially the last paragraph) and others before him (he cites Chatto, 1848). Even the proponents of Egyptian origin (ridiculed by Waite) did not deny this use as central; it was by encoding their wisdom in the form of a game, they maintained, that the Egyptians showed their wisdom best, by giving what was so encoded the best possible medium for its survival (de Gebelin, Article 1, first section). It is also obvious that such objects as cards would have had other uses, sooner or later. It is only the issue of what use when that is at issue.
TRUTH AND THE PAST
Dummett's argument against "occult" uses in
Game of Tarot is the lack of evidence for them. What is the relationship of facts about the past to evidence now for them? But we are tempted to say: surely there are facts in the past for which there is currently no evidence! On this topic, what is relevant is Dummett's theory of truth as he applies it to statements--"assertions", he calls them, sentences used to convey information--about the past. Assertions about the early tarot seem to me a subset of this larger category.
Dummett's philosophical account of such statements seems to me of possible relevance to some of our differences on this Forum. History is about the uncovering of historical truth. But what are the truths of tarot history? Is the quest for truth only the uncovering of solid, documented information about the cards and the use or uses to which they were put, or can it include speculation about decks of various possible numbers of trumps, of unspoken symbolic meanings, drawing inferences from the various symbol systems of the times, and unspoken uses such as divination, again by way of inference from divination and divination-like practices of the times?
Looking at Dummett's writings on truth and the past, what I find is that he held different views at different times. He says as much in the 2005 preface to his last published book,
Thought and Reality (based on his Gifford Lectures of 1996). First, he identifies himself as holding a "justificationist" theory of meaning: to know the meaning of a sentence is to know what would justify its utterance. Justification at the time and place indicated by the sentence would thus qualify a vast number of sentences as meaningful. But then:
All turns on what notion of truth is appropriate to a justificationist theory of meaning. The question has worried me for many years. For a time I had believed that the justificationist must be an anti-realist abut the past, a conclusion about about which I always felt uneasy. The Gifford Lectures firmly repudiate this; but the conception of truth that they propose does not make so conciliatory an advance in a realist direction as does that proposed in the Dewey Lectures.
He says that between the two (the Gifford Lectures of 1996 and the Dewey Lectures of 2001 or 2002), he does not know which is the right conception. He also says that the Gifford Lectures represent "a transitional phase in my thinking about the subject. Chapters 5-7 express views I no longer hold". In other words, he disagrees with the Gifford Lectures. Well, I guess a logician is free to contradict himself; probably he is honestly presenting two sides of his thinking, not wholly consistent with each other.
In any case, he has held, at one point or other, three conceptions. First is anti-realism. This is the view that an assertion is true just in case there are memories or evidence that justify its assertion. In the 1970s (e.g. "What is a theory of meaning?" 1976), the term he used was "verification"; "justification" is a little broader than that, including more than just "evidence": memories are not a type of evidence, he rightly says. It means having reasons sufficient to justify making the assertion.
The theory of truth opposed, in his view, to "justificationism" is called "truth conditional": an assertion is true just in case it satisfies the conditions that make it true. These have nothing to do with how the person making the assertions could justify making it, but rather whether what is asserted corresponds to reality. Frege and early Wittgenstein held that theory, as do a majority of analytical philosophers today, Dummett says.
In relation to tarot history, according to the "anti-realist" version of "justificationism", if there is no evidence or memory-reports of divination with tarot early on, or its use to encode "occult" meanings, then that is that. Such speculations cannot be part of the tarot's past, unless sufficient evidence is later found. And they are only meaningful to the extent that the person entertaining them has a clear idea and decision procedure for what would count as evidence for and against.
On a truth-conditional theory of truth, however, speculations not grounded on evidence can still be entertained as hypotheses even if the person entertaining them has no idea what would justify adopting them. What he has to know is whatever would make them true, not what would justify their assertion. For example, "in the 15th century, tarot was used for fortune-telling based on the titles of five trumps chosen at random" would be true if in fact tarot was used for that purpose. The sentence expresses a "thought" (Gedanke, Frege's term) or "possible situation" (or "possible state of affairs" in another translation: mögliche Sachlage; also "possible fact", mögliche Sachverhalt, early Wittgenstein's terms,
http://people.umass.edu/phil335-klement-2/tlp/tlp.pdf) which, if it corresponds to an actual fact, is true, and false otherwise.
The problem with "truth-conditional" theories of truth, and meaning, is that they have what seems to be an unavoidable circularity. Dummett says (p. 78 of the Gifford Lectures);
A blanket account of understanding a statement as knowing what it is for it to be true is useless, because circular: it attempts to explain what it is to grasp a thought in terms of having a thought about that thought.
(I am not with this quote trying to give Dummett's whole argument, but merely documenting his conclusion.)
Perhaps it was to avoid a similar circularity that Dummett switched from defining truth in terms of "verification" to that of "justification", in as much as "verification" just means "establish as true" (as the Latin root "ver" implies). "Justified" does not mean, for Dummett's class of sentences called "assertions", established as true. It means that for the assertion there is a decision-procedure that will result in establishing it as true. Knowing the decision procedure is what is involved in using the statement "correctly", as Dummett puts it. This language of "justification" applies equally to statements of the form "there is objective evidence that P" and "I remember that P".
In any case, the "anti-realist" view, Dummett pointed out even in 1969 (in an essay called "The Reality of the Past", reprinted in his book
Truth and Other Enigmas), has consequences that, if not circular, are at least very uncomfortable. If it is true that Dummett is wearing a red tie in 1996 as he gives his lecture, it should be true exactly a year from then that exactly a year prior he was wearing a red tie. But if there is no evidence, or no one has a memory to that effect, then on the anti-realist view, it isn't true that he was wearing a red tie. Truths about the past will fade away over time, as evidence and memory disappear. There is no contradiction here, despite appearances, just a discomfort. It is one that Dummett appears to have been able to live with, given his commitment to justificationism.
But in 1996 he decided that a justificationist can safely repudiate anti-realism about the past. A statement that is justified at a particular time remains true by virtue of its being justified then:
In the Gifford Lectures, a proposition is reckoned to be true just in case we, as we are or were, are or were in a position to establish it to hold good;...
In the text (p. 78), he gives a rare example:
The number of people who witnessed the execution of Charles I must have been odd or even, although we cannot now say which.
This is a case where in the present we can say that a crowd has to be odd or even in number and can determine which it is; so the disjunction applied to the past time is a true proposition now.
In his 2002 Dewey Lectures, Dummett goes on to say, he again revised his view about when a statement about the past is true. He says (I am still quoting from his 2005 preface to his 1996 lectures, in fact from the same sentence I just quoted):
...my present standpoint, as stated in the Dewey Lectures, is that it is true just in case anyone suitably placed in time and space would be or have been in such a position. The difference has an evidently far-reaching effect: far more propositions will be rendered true under the Dewey than under the Gifford conception. For instance, the colour of a flower must be determinate, since such an observer would necessarily have evidence for it, even if such evidence is not available to us.
However what color it is, is at present still unknown. That it has a determinate color is a matter of the flower's reflecting, or not, particular wavelengths of light at the earlier time, and not merely the truth of a subjunctive conditional, "If someone had looked in the right place, he would have seen a flower of determinate color" (as he clarifies in his 2005 article, "The Justificationist Response to a Realist").
What the situation is if it is a tie that Dummett randomly chose in the dark, then covered with his overcoat all day until it got dark again, and then put it back on the rack with the others, he does not say. Presumably no determinate color can be attributed to it, even at the time he is wearing it, since there are no light-waves to be had. This seems to me quite odd. In the present, we can judge whether Dummett is wearing a red tie in such a case by having him show it to us. The same should be true of his past wearing of a red tie: it is true that he was wearing a red tie if unbuttoning his overcoat would have revealed one. It seems to me that it is precisely the truth of such "If...then" statements that constitutes the reality of the past--and by the same reasoning, the future, too (if his overcoat is unbuttoned on the future date, we will see then that he has a red tie).
The relevance to tarot history is that the truth of, say, "The inventor of the tarot had in mind occult meanings" is not just a matter of whether that person told anybody that that was what he had in mind, such that a suitably placed observer could have overheard, but that if a suitably placed (i.e. trusted) observer had asked him if he had occult meanings in mind, he would have said yes and, if asked, justified himself or herself. That is how we would find out in the present. The same would seem to apply to the past.
I am not clear on just what the difference is between Dummett's 1996 and 2002 formulations,. It would seem that he made his first concession to realism just so as to deal with the awkwardness of Dummett's red tie and other conditions that he raises, such as the continuing validity of inferences when moved from the present to the past. Perhaps the 2002 formulation has to do with the type of means that can render a statement true. For example, there may in the future be advances in logic or science that could justify something, advances that the speaker now has no knowledge of (e.g. a proof of the existence of God as the all-good creator of the universe, of which Dummett says there is no proof at present but hopefully will be in the future. as early as the time of "our great-grandchildren", p. 151 of
The Nature and Future of Philosophy, originally published in Italian 2001). In such a case, the meaning of a sentence remains its most direct method of justification, but it may be one unknown to the speaker. But I do not see what this consideration has to do with the determinateness of the color of a flower.
In either formulation, the past does not disappear when evidence disappears; it merely, from our later perspective, loses its details, those that cannot be justified in the present (the determinateness of its colors, the oddness or evenness of its numbers); it retains what can be inferred from the present, applying it to what a properly situated observer could have seen in the past. And regardless of what can be inferred in the present, it retains whatever a properly situated observer in the past could have seen then.
This seems to me an important shift in Dummett's thinking. I think it applies also not only to what can be established--fully justified, fully reasoned--but also to what can be somewhat justified, given some reason for believing. (Dummett explicitly allows generalization to probabilistic reasoning; in fact, that is one basis, in his 2005 article, for distinguishing empirical from mathematical statements.) So it is to be expected that inferences from known facts about tarot decks to unknown ones would be only partially established, and otherwise specified by disjunctions of possibilities, like the odd or even number of persons at Charles I's execution, or the presence of some color, but we don't know which. In Dummett's later thought, we can go beyond hard evidence into shadowy realms of sketchy alternatives. It is in this way, I think, that it is preferable to give a series of disjunctions of reasonable possibilities for the earliest numbers of trumps (as I did on this thread at
viewtopic.php?f=11&t=975&p=14457&hilit= ... ipt#p14457) rather than the one number that is documented decades after the tarot was invented.
The effect of Dummett's shift in perspective might be present in
Wicked Pack of Cards, where he seems more tolerant of the possibility of systematic "occult meanings" in the tarot than he was in 1980, at least as I interpret the passage on pp. 33-34, quoted by M. J. Hurst at
viewtopic.php?f=11&t=975&p=14304&hilit=occult#p14304. Dummett also is willing to say that Etteilla's tarot divination system is "largely fed from older French cartomancy" (p. 94), without insisting that it is a cartomancy invented by Etteilla himself earlier, something not tolerated in 1980. Whether this shift affects his writings on tarot after 2001, or even his book in Italian of 1993, I do not know. His 2006 article on the second-artist cards of the PMB certainly seems more speculative than anything else he has written. I see nothing wrong with that; he was merely more willing to toy with possibilities than he was earlier.
Another thought: since Dummett never was very happy with anti-realism, it may be that the shift in theory has more importance for us, in our implicit theories dealing with the past, than it did for him.
In the 2002 Dewey Lectures (
Truth and the Past) he says something that seems to me dangerously close to a truth-conditional theory of meaning. What justifies a statement at the time of optimal justification is given by "the composition of the sentence and the way the meanings of the words that make it up are given" (p. 53; he calls this "canonical justification" as opposed to indirect, inferential justification). But that is very much like what Frege and the early Wittgenstein said was what determined the sense of a sentence; and from that one could determine what it was for the sentence to be true, what early Wittgenstein called a "possible fact" (mögliche Sachverhalt, or Sachlage). It is from the words of "the cat is on the mat" and their order that we know what it is for that sentence to be true; i.e. it is true if the cat is on the mat, if that possibility is a reality. The apparent circularity is broken by our being able to justify not that sentence but rather that we know the meanings of the individual words as they combine into comparable sentences. This is justified by our ability to use the words correctly in other sentences that are true, including ones we have never heard before. It is a justificationist grounding for a truth-conditional theory of meaning.
We show that we understand words, as Dummett says, by using them correctly, i.e. with justification, in different true (or apparently true) sentences. A tie is red, a bowl is red, a ribbon is red, and so on. Besides sentences we have been taught are true, we can make up sentences on our own that others competent in the language judge to be correct. They are justified because the objects to which we attribute redness look red to us; more than that, other people generally agree that we have made the identification correctly. And more than that, such identifications are of practical value in our lives. If I spilled gravy on my red tie yesterday, its redness tells me to look carefully before I wear a red tie today, more carefully than I would at other ties I own. This is justificationism, pragmatism, and truth-conditionalism all in one (as pertaining to the justification for our understanding of particular words, I want to emphasize, not to our new uses of the same words once we have demonstrated that we understand them). But as a result, our words can never be used assertorically (if there is such a real category) in a meaningful way except in the context of our experience as language-users in ways we can justify to ourselves and each another.
I do not want to make much of these last observations (of the last two paragraphs). They may have little relevance to the activity of tarot history. Dummett's later thoughts on the reality of the past, however, are a different matter. In Dummett's later theory of truth, as opposed to the earlier, is that in the later we can put forward inferences from present facts to disjunctions of sketchy alternatives, for which in turn there can be developed a variety of considerations for and against. Paradoxically, while we can say more in this later theory, we know less. In the earlier theory we did not have to concern ourselves with Dummett's tie at all; if there was no evidence or memory of it, we could disregard it. In the later, we are faced with our lack of knowledge: if he was displaying a tie, it had a color, but we don't know which. But from that lack of knowledge may develop new knowledge, good inferences we would not have found otherwise.