Keywords

1 Introduction: Not Just Any Old Bird

Our bird, a cockatiel named Hedwig (hatched, purchased, and named well before the invention of Harry Potter’s owl), died last summer at the honorable age of 23. It was a loss to the whole family, but for our youngest daughter (aged 2) this was a really hard blow. We did not really notice her strong attachment to the bird in the beginning. We knew she loved animals—all animals—more than most kids do. She could stare at, and proclaim her love to, a calf, an ant, a dog, a toad, a cat, and, of course, to Hedwig. But she could also call any bird “Hedwig.” So did she really know Hedwig? Or did she think that all birds were called “Hedwig” (and not “bird”). Would it be fair to say that she had confused type with token, generality with particularity? She had learned something, that’s for sure. She did not, for example , call cars or trees, or the food on her plate “Hedwig.” She did not call airplanes “Hedwig” (and I have seen kids call airplanes “birds” and parents not correcting them on the basis that it was “close enough”).

But nearly half a year after the birds passing, our two-year old happened to watch some photographs in which Hedwig figured in the background . And tears started to roll down her cheeks. “Where’s Hedwig now?” she asked. That Hedwig really was a unique individual to her became quite obvious. But this also means that some of the uncertainties that we have had regarding her use of Hedwig’s name appeared in a new light. It was quite clear that her care for Hedwig was very real and that Hedwig was not just any old bird. Of course, we had corrected her earlier on when she had used Hedwig’s name too inclusively: “No, that’s not Hedwig. Hedwig is at home. That is another bird.” And we had also—of course, all parents do, all parents must do—trained her using ostensive definitions in a very traditional sense, saying things like “That’s Hedwig” while pointing at the bird.

In hindsight, these kinds of corrections now sound a bit odd. For they may make it appear precisely as if our daughter actually had confused token and type, particularity and generality. And clearly that was not the case. But we might say, I take it, that we did teach her that there is a kind of distinction to be drawn here, at the same time as we taught her what a name is. We never thought about it that way (“Hey, should we practice the type/token distinction today?”) at the time. We know that now only when we have stopped talking about it. Teaching was done when she talked about these things in a way that did not spur correction, and one can clearly see, in hindsight, that we had “taught” her much more than we thought we did.

This, so it seems to me, suggests that it is quite hard to point out a particular moment when our daughter had learned such a straightforward case of language use as “learning the name of one’s pet bird.” And it also shows that one of the reasons why that may appear difficult to do is because it is surprisingly hard to say what a child has learned when she has learned to use a name properly—and so one may say that it is hard to pinpoint what is to be included in this “properly.” (Perhaps one may also say that coming to know Hedwig—who she was, what that particular name meant, also included learning that birds are living creatures and that all living creatures eventually dies?)

So, how do we know when learning has taken place? When is a teacher’s job done? One answer that may be drawn from Wittgenstein’s work is: when the student is able to go on alone (see, e.g., PI, §§143–151).Footnote 1 But then one may ask: “What does the pupil know, when he or she knows how to go on, on his or her own?” And now several responses seem natural to resort to. For example : The pupil knows how to go on when he or she knows what a word/concept really represents, because only then would he/she be able to use it without guidance, and sense would be externally secured. Another kind of answer that one may naturally reach for is to say that a child has learned how to go on alone when she has grasped the regularity underlying the phenomena at hand. We know how to use a word in new contexts when we know what it means, or we know how to use the words we have learned when we know the rules that guide their correct use. Something is not right in such responses. And I will show why I think so in what follows. Indeed, I even think it is fair to say that Wittgenstein’s talk about “learning how to go on” and “knowing how to go on” is employed precisely to debunk the hollowness of philosophical attempts to explain “knowing the meaning ” in terms of “knowing what name connects to which thing” or “knowing which rule to apply.”

More specifically, I will suggest that a focus on rules of language runs the risk of misrepresenting the real nature of the linguistic regularities (which may also be called normative regularities) they are meant to explain or elucidate.

2 From Names, to Rules, to Life

It would be bizarre to suggest that normativity is not involved in language acquisition . And it would be nonsensical to say that there are no regularities to be found in language use. But are we then to conclude that the regularities we find in language use, and the kind of normativity we impose on each other as we continuously try to learn our languages, must fall back on some linguistic rules that guide and control our uses? Would such an account enable us to think clearly about scenes of instruction where we speak to a child, agree with her projections of words, and then, in hindsight, revise out judgments about what it was that the child knew (as was the case with Hedwig and our daughter)?

There is a ladder to be climbed here I think. The first rung of the ladder is the assumption that a child knows a word when she knows what the word stands for—words are names for things, the meaning of a word is what the word “stands for” or “refers to.” This—clearly one of the most persistent and tempting philosophical thoughts—is a misguided way of thinking about language on a larger scale. Knowing how to use and understand the name “Hedwig” was, for our two-year-old, knowing much more than knowing what our bird was called. It included, for example , knowing that Hedwig was living but is now dead, what death is, that there is a difference between a particular individual and a general kind, i.e., that not all birds are called “Hedwig,” and that calling one’s newly attained toy-bird with batteries inside “Hedwig” can upset one’s older sister because that name “was taken” and because it is disrespectful to call a toy by the name of one’s newly diseased animal. Of course, these remarks can by no means, nor are they intended to, challenge the idea that a words are often names for things. That would be downright stupid. Words refer too. But it is precisely because we often can tie a string between a word and a thing, and let a particular light shine on that string in philosophy books that tempts us to think that this is all there is to it, or at least think that this is the basic relation that we need to think about.

The second rung of the ladder would be to say that linguistic regularity cannot be explained as a name—thing relation only. It is also probably quite uncontroversial to say that a shift has occurred from words to sentences in philosophy of language . We also need a rule which tells us when, in what circumstances , a particular word-thing relation holds and is relevant. That is, linguistic sense and regularity, as well as the normative force one may have as a teacher, fall back on a specific linguistic, grammatical, rules which guide and control all our uses. The rule is what assures us that this projection of our words is OK, as well warrants us in correcting somebody who is using (or attempting to use?) language in an outlandish, or too eccentric, way.

This line of reasoning is very familiar today, and many philosophers who underline the importance of guiding, normative, rules are also philosophers who proclaim to be followers of, or at least inspired by, Wittgenstein . Some of the earliest and most influential and formative interpreters of Wittgenstein have also placed “rules of language” at the center of Wittgenstein’s philosophy . Crispin Wright, for example , claims that “To know the meaning of an expression is to know, perhaps unreflectively, how to apprise use of it; it is to know a set of constraints to which such uses must conform” (Wright 1993, p. 24). Tim Thornton suggests that “The correct (and incorrect) applications [of a word] are determined by the rules that prescribe its correct usage” (Thornton 1998, p. 32). Hans-Johann Glock suggested that “linguistic understanding involves mastery of techniques concerning the application of rules” (Glock 1996, p. 223), and he has also suggested that the point with Wittgenstein’s philosophy is that it “clarifies grammar, the set of rules by which we determine the correct use of words” (Glock 1991, p. 70). Peter Hacker claims that “Possessing a concept involves being able to use a word in accordance with a standard of correctness” (Hacker 1972, p. 219).

But then we should ask: What were the rules that we (supposedly) had taught our daughter? And what, more concretely, does it mean to teach a child a rule for the uses of our words without, or at least before, one teaches her the word? This is a bit bantering, I know. Nobody can think that we teach our children a rule before we teach them the word. But this kind of ragging remarks, the possibility of them, point to something too. For if the regularity itself, the standard of correctness, cannot be so much as gestured at without using the words they are meant to guide and control, then why should we say that the one must precede the other? And can one really say that our uses of language are guided by an underlying rule if we cannot even discern that kind of regularity without looking at the word in use, already in use?

Wittgenstein can be said to climb these two rungs of the ladder in the Philosophical Investigations . The idea that words are names is indeed the very first thought that Wittgenstein opens his Investigations with. And it is fair to say that Wittgenstein’s builders (however else one may understand them and the point of the example) seek to destabilize the certainty with which one claims that words are names for things. As soon as the builders are said to formulate sentences, combination of words, the idea that meaning can be reduced to reference crumbles. It is not that Wittgenstein wants to deny that words name things—as if one could deny the fact of names altogether—but it becomes evident that we will not get the regularity of language in view by means of explicating meaning in terms of reference. Reference alone cannot give us a way to understand the manifold ways on which just a few numbers of words can be used.

But how many kinds of sentences are there? Say assertion, question, and command?—There are countless kinds: countless different kinds of us 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 obsolete and forgotten. (We can get a rough picture of this from the changes in mathematics ). (PI §23)

Just as one may be tempted to think about “learning to speak about Hedwig” as a question about learning to make a connection between an individual animal and a name, it may be tempting to think that “learning language consist in giving names to objects . Viz, to human beings, to shapes, to colours, to pains, to moods, to numbers, etc.” (PI §26). But that, Wittgenstein quickly notes, merely appears to get at something that is “preparatory to the use of a word” (PI §26). The pointing, the ostensive definition—“This is Hedwig”—has its contexts too: “the ostensive definition explains the use—the meaning —of the word when the overall role of the word in language is clear” (PI §26). Names, ostensive definitions, talk about reference are parts of language, but “the speaking of language is part of an activity, or of a form of life” (PI §23).

Thereby, Wittgenstein leads us to think that if we want to understand what “coming to know the meaning of a word” means (where it happens, and how, and why), and what one has learned when one has mastered a concept, we need to look at some other level. And, it is not surprising that “rules ” came to take a central position in such a quest—for Wittgenstein does talk a great deal about rules , and where else are we to look if we want an account of the normativity of language if not there? There has to be a way of determining what one has to know when one knows how to use a word correctly! Right? And even if one were to be unable to spell out exactly what the rule was, there still has to be a kind of “standard of correctness,” as Hacker called it to fall back upon when a teacher says to a student “repeat after me” or “go on like this….”—right?

Wittgenstein now asks us to think about what it means to teach somebody how to “write down a series of signs according to a certain formation rule” (PI §143). The series Wittgenstein picks is perhaps the most simple we can think of: “the natural numbers in decimal notation” (PI §143). Quite likely, the first thing that happens is that the teacher tries to teach our student how to copy the numbers. Importantly, Wittgenstein remarks that “the possibility of getting him to understand will depend on his going on to write it independently” (PI §143). Whether or not the teaching continues depends on the pupil’s reaction . If he continues the series the correct way, like we do, teaching may come to an end; if not, not. But how many times does the pupil have to continue the series in the same way as we do, in order for us to say that the pupil actually got it? “Clearly,” Wittgenstein remarks, “you cannot state a limit here” (PI §145). Clearly? Really? Is there no “it” we can assume that the pupil knows? This is the point at which it seems natural to claim that it is the rule itself, or the system, that the pupil must have mastered.

Suppose I now ask: “Has he understood the system when he continues the series to the hundredth place?” Or—if I should not speak of ‘understanding ’ in connection with our primitive language-game : Has he got the system if he continues the series correctly so far?—Perhaps you will say here: to have got the system (or, again, to understand it) can’t consist in continuing the series up to this or that number: that is only applying one’s understanding . The understanding itself is a state which is the source of the correct use. (PI §146)

There are two central lines of thought here that Wittgenstein criticizes, or challenges. The first is how natural it may seem to think of the application of a rule as something quite different from the rule itself. Indeed, it seems almost self-evident. Here is the rule. Here is me applying it. That the exemplary case here is “natural numbers in decimal form” further underlines the reasonableness of such a reaction . “For the series is infinite and the bit of it that I can have developed finite” (PI §147). The second thing worth emphasizing here is that understanding also (and expectedly?) is divorced from the “application,” and said to be residing in some kind of relation to, or attitude toward, or insight in, the system, the regularity, itself.

But wait a minute! Is this not precisely the image that underlies the very idea that language use must be guided by means of a set of underlying rules? Let us look again at these classical formulations of Wittgenstein’s so-called rule-conception of language:

To know the meaning of an expression is to know (…) a set of constraints to which such uses must conform. (Wright 1993, p. 24)

The correct (and incorrect) applications [of a word] are determined by the rules that prescribe its correct usage. (Thornton 1998, p. 32)

[L]inguistic understanding involves mastery of techniques concerning the application of rules. (Glock 1996, p. 223)

Possessing a concept involves being able to use a word in accordance with a standard of correctness. (Hacker 1972, p. 219)

Are not these formulations clear expressions of the idea that the application of a rule is something quite different from the rule itself, and the idea that understanding must be sought for in the system, the regularity, itself? And, are not these merely alternative formulations of the idea that “The understanding itself is a state which is the source of the correct use”? It seems as if these followers of Wittgenstein are rehearsing the very image that Wittgenstein aims to challenge—for example , when he somewhat mockingly asks: “But what does this knowledge consist in? Let me ask: When do you know the application? Always? day and night? Or only when you are actually thinking of the rule? do you know it, that is, the same way as you know the alphabet and the multiplication table? Or is what you call ‘knowledge’ a state of consciousness or a process—say a thought of something, or the like?” (PI §148).

Wittgenstein aims to show that “knowing how to go on” was already, as it were, the full answer to the question what it is one must know in order to follow the lead of one’s others. “The grammar of the word ‘knows’ is evidently closer to that of ‘can’, ‘is able to’. But it is also related to that of ‘understands’. (Mastery of a technique.)” (PI §150).

What has gone missing, one may say, is precisely the life of language —the fact that “the speaking of language is part of an activity, or of a form of life” (PI §23). I remarked above that a shift has occurred from words to sentences in philosophy of language. There are clear connections to make between the focus on words with an interest in reference and the representational character of language , and the focus on sentences and the interest in (grammatical, logical ) rules . These together would mark the first two rungs of the ladder of this paper. The third rung of the ladder is the attempt to return to the rough ground, as it were, and look for linguistic regularity in practice, as lived.

3 The Life (of “&C.”)

I want to suggest that one problem here, one reason why we may be tempted to cling on to the “rule-rung” of the ladder mentioned above a bit too long, is partly due to Wittgenstein himself. The way I see it, Wittgenstein chose simple mathematical series as exemplary for his discussions, precisely because these kinds of “linguistic” regularities are so tempting to think of in terms of guiding, externally existing, rules —almost Platonic (in the pejorative sense of the “term”…). But it appears as if many of Wittgenstein’s interpreters think that his penchant for simple mathematics was due to Wittgenstein wanting us to think of the regularities of our language as just a variation of a simplistic mathematical rule (as if Wittgenstein had not gone to great length in trying to show that mathematics too was far from as simplistic as many philosophers are prone to think).

So it may be helpful here to think more about kinds of regularities in language use that are not so self-evidently “rule-like” in order to further make clear that regularity in language use, the normative force of language, does not depend on, or fall back upon, a kind of rule, or form of language, that precedes all articulations (correct and incorrect).

One clear, common, colloquial, everyday, image of “knowing how to go on” (that does not immediately make us think about underlying or guiding rules ) can be found in the simple phrase “&c.” “&c.” is the old abbreviation of etcetera. The “&” refers to “et” which is the Latin for “and,” and the “c” stands for “ceteri”—i.e., “others.” Literally then, the Latin et cetera means something like “and the rest of such things,” and we usually use the phrase “etcetera” when we want to suggest that a list that has just been started should go on, be made longer (but probably not ad infinitum). And we may also use “etc.” to signal that there is no need to give the list in full—when the rest of the list is, as it were, too obvious, or too tedious to repeat. Et cetera thus means something like “go on in this manner” (where “in this manner” means something like “like this, but do not merely repeat the same over and over again”) and “go on for some time, but not for ever—I won’t bother specifying it, you’ll get it.” So there is an instruction and a following of the instruction involved, but both the instruction and the “carrying out” may appear rather vague. This is the kind of phrase we employ when we trust that the other will be able to go on alone (without my guidance and supervision).

&c. may also be used in statements that requires that we “fill in the blanks”—as in “They lost everything in the fire; house, boat, both cars, photo albums, etc.” The “go on like this” then, the “rest of such things” implied by the “&c.,” need not be “strict” instances of the same concept (as in “I like berries, like blueberry, cloudberry, lingonberry, etc.”), but are quite often more loosely ordered: “… and by ‘original’ I mean the text which you read or copy; the dictation from which you write; the score from which you play; etc.” (PI §162); or “The adverbs that can be inserted in ‘How … do you know?’ are few in number and even fewer in classes. There is practically no overlap with those that can be inserted in ‘How … do you believe?’ (firmly, sincerely, genuinely, etc.)” (Austin 1979, pp. 81 f.n. 2), or “Either I myself may be dreaming, or in delirium, or under the influence of mescal, &c.: or else the item may be stuffed, painted, dummy, artificial, trick, freak, toy, assumed, feigned, &c.: or else again there’s an uncertainty (it’s left open) whether I am to blame or it is—mirages, mirror images, odd lightening effects, &c.” (Austin 1979, p. 87). What follows from the “etcetera” is, as it were, “left open,” but it would be wrong to say that this signals an uncertainty. There is no obvious way, not one way only, to go on here, yet we know what it would mean to go on, and also what it would mean not to go on.

Etcetera is thus an expression we use in order to signal that there is a form of regularity about our shared life with words here, and we are counting on our others to see that. And we use it (“etc.” that is) without hesitation. So, in a sense, it is an expression of trust. Only that “trust” often becomes relevant, something to relate to, in cases where the trust is questioned, or not easily discerned, in a similar way as “certainty” often comes together with doubt. (I do not trust you to not stab me with a knife; I am not certain that the ground on which I walk will carry my weight. But if you “inform me” that you are not going to stab me with a knife, or ask me if I think that the ground I am about to walk will carry my weight, doubt, and evaluations of my surrounding may be spurred, and then certainty and trust may be expressed.)

Focusing now on the kind of trust (the kind of unwittingly held expectations, or silently expressed acknowledgments of a shared horizon) that is involved in cases where we say “etc.” or “and so on” should broaden our horizon and widen our image of what “understanding ” and “knowing” may mean if, say, a pupil follows the lead of his or her teacher.

Compare these two scenarios: First, a teacher asks her first graders to “go on like me” after she has said “red, orange, yellow… and so on.” One of the kids raises his or her hand and says, “black, blue, white, and pink.” The teacher says “Good!” A few days later, the same teacher asks her students if they know how to “go on like me” after she has said “red, orange, yellow… and so on.” And one of the kids raises his or her hand and says “black, blue, white, and pink.” Some of the kids giggle. The teacher says “No. That is not right, I’m afraid. Anyone else?”

What can possibly have happened? What has gone wrong? The scenes appear to be identical, and the persons involved are the same. The collapse of sense here is due to a lack of understanding of context , or of the situation. If we imagine these two scenes as really being identical, it is true that something utterly strange is going on—so strange, indeed, that we are likely to disregard this as too unreal, nonsensical perhaps.

But suppose, instead, that in the first case, the children were practicing colors—how many there are, what they are called, what they look like, how they are spelled (etc.). If that is the case, it is quite clear that the child did “go on by herself” in exactly the right way. And suppose now that the second scenario—in which the (same) child raised his or her hand and responded with the same words to the same question but got it wrong—happened in a class where our poor child was fooled to believe that they were discussing “colors” (it all looked the same, and his or her teacher did say the same things as last week), because she just got back from a doctor’s appointment and so arrived at class too late (just at the moment at which their teacher popped the question) and did not know that the topic under discussion was physical phenomena like rainbows (the color of them, for example , and the order of the colors in them). If this was the case, there is nothing strange with the teacher correcting the child. It just was the wrong answer!

Thus, this it is not merely a case of us, being external observers, failing to understand the context of these two scenarios. The student who got it wrong in the second scenario did not understand the setting either. And I am inclined to say that that is why she did not know how to go on in the right, expected way. That is, if a child fails to complete a series, or does not manage to “go on like the teacher,” follow his or her lead, this is not necessarily because the child does not understand the rule, or some other underlying normative structure, of this particular language -game . It is more likely that the child lacks the relevant kind of preconception of the order of things, does not understand the horizon from which the example originates, and does not share the world with the teacher. This is one of the reasons why I think that Stanley Cavell points to something of great importance when he remarks that “In learning a language, you do not merely learn the pronunciation of sounds, and their grammatical orders, but the ‘forms of life’ which make those sounds the words they are, do what they do” (Cavell 1979, p. 177). But that also means that Wittgenstein must be seen as steering our attention away from the idea of language as rule-governed, toward a more holistic way to bring linguistic regularity into view when it is called for, and if it is.

This last comment about when an attempt to bring linguistic regularities into view is called for, if it is, is prompted by a worry that philosophical questions that call for such responses tend to enter the world in which language us lived sideways on. A typically philosophical posture here would be to say that “Meaning does not depend on rules , but on the form of life in which speaking takes place. What is more: what cases such as the two series of colours show, meaning is not only context sensitive, but occasion sensitive” as Charles Travis may be said to suggest (Cf. Travis 2008). Now, I think Travisian reflections are true and very helpful on many occasions, but as a general theory of language , I am inclined to say that they must miss the target. If it is wrong to say that rules are what bring meaning to language , and it is problematic to say that contexts are what invest our signs with meaning, it is also confused to say that meaning always is invested in our words “occasionally”… That is, as a source of philosophical criticism, Travis’ reflections on occasion sensitivity are extremely helpful, but if we think of these reflections as forming a general theory of language , then “occasion sensitivity” comes in too late, as it were. This would be a theoretician’s response to a puzzle and that would be its only function. “Occasion sensitivity” may be said to describe features in conversational scenes, but it does not really play an actual or guiding role in conversational situations, as if one would have to consult one’s own faculty of occasion sensitivity before one “decoded” an utterance. Also, in scenes of instruction, scenes in which language is learned, we do not teach our pupils an almost infinitely long list of occasions, and a complementary list of “suitable” ways of turning ordinary words in them (and I am quite certain that Travis in agreement with me here).

It may be right to say that the child who failed to go on listing the colors of the rainbow did not understand the context of the utterance and was thereby not occasion sensitive enough. It would be wrong, however, to say that she did not understand the words of the language , or the sentences. But it would also be strange to say that she understood the words of the language, the sentences, rightly. What she did not understand was the context. And it is true that words belong in contexts . But we should be very hesitant to introduce either “context” or “occasions” as “suitable alternative responses” to the philosopher that proclaims: “There has to be a way of determining what one has to know when one knows how to use a word correctly!”

Contexts and forms of life are not answers to questions about what invests words with meaning . But that is not to say that contexts are irrelevant, or disconnected from the sense of our words. Quite the contrary actually. So John Searle, for example , has a point when he remarks that “cut” means quite different things in “cut the grass” and “cut the cake” (Searle 1980, 1994). And therefore, it may appear to make perfect sense to say that the lawnmower is “implied” in the first, and a pair of scissors are “implied” in the second. But this does not mean that one cannot “cut the grass” with a pair of scissors. Not only in the sense that one can use a pair of scissors to cut grass, but because “cut the grass” may be used in a great many varieties of ways too. This particular sentence is, for example slang for trimming one’s pubic hair. So I want to say that even though it is true that one wouldn’t be cutting the cake if one ran it over with lawnmower (Searle 1994, p. 640) and it is true that it would look odd, untoward, and unexpected to see somebody try to trim say, a soccer field, with a pair of scissors; just like one can feed peanuts to a monkey but not pennies, and pennies to a parking meter but not peanuts (Cavell 1979, p. 183); there is still something with Searle’s view that does not seem right. Searle’s view suggests that how one ought to do it (with a pair of scissors or with a lawnmower) follows from the semantic content of the sentence. I want to say that the idea of semantic content cannot be disconnected from sentences used in concrete situations, which means that the whole idea of “semantic content” (or “sentence meaning,” for that matter) becomes deeply problematic. Contexts do not bring meaning to words and sentences; words and sentences mean in contexts. We simply do not know enough about the world in order to determine what “cut the grass” means (what semantic content it supposedly has) without knowing what, if anything, we are doing with our words. And, after all, to understand a context is not to understand the meaning of a word, or the root of all meaning, but to understand the context. This can be seen as variation of Cora Diamond’s remark: “To give an account of meaning in terms of assertion conditions is to remain with our eyes fixed in the wrong direction” (Diamond 1990, p. 15).

4 Pedagogical Postscript on Hedwig and the Question of Parroting

When our two-year-old learned talk about Hedwig, she learned to employ the pet’s name in a great number of contexts and to include talk about Hedwig in a great number of activities. To learn to understand, as it were, what “Hedwig” denoted in this case, included learning a great deal about life and death, sisterhood, the difference between a pet and toy, and, perhaps also, to come to see where the philosophical–theoretical description of language use captured in the type/token comes from, and so on and so forth. So there is a sense in which it is true to say that to merely repeat the birds name, pure parroting, is not to know the name.

Teaching and learning need to be brought home to our shared practices . And there can be something dubious about abstracting principles and summarized images of expected “learning outcomes” etc., not because we cannot say, or hope, or predict, that our pupils will learn this or that, but because such abstractions tend to misrepresent the objects . We do not, for example , cook together (in school or at home) in order to let our children see or learn about a practical application of the principles behind measuring (say, ounces and cups, or deciliters and centiliters, even though they will nevertheless learn what ounces, cups, or deciliters and centiliters are; and one thing that they will learn is that ounces and cups, or deciliters and centiliters are not abstract principles). Practical learning experiences are not practical ways to learn about theoretical entities; they are, as it were, the thing itself. This is why I said that “knowing how to go on” is already, as it were, the full answer to the question what it is one must know in order to follow the lead of one’s others.

“To learn to speak” may be described as learning to use language in a way that adheres to the regularities we find in language. But it would be wrong to say that this amounts to the same thing as “learning to apply a rule.” Rush Rhees thus put his finger on something immensely important when he said: “Show how rules of grammar are rules of the lives in which there is language ” (Rhees 1970, p. 1970; Cf. Diamond 1990). Learning is a matter of coming closer to one’s others, to one’s community. But, of course, given that learning means “learning how to go on, on one’s own,” it follows too that becoming a member of one’s community is never a matter of parroting.Footnote 2