Version classiqueVersion mobile

Greek and Latin Expressions of Meaning

 | 
Andreas T. Zanker

Chapter 4: Metaphor, Polysemy, and Meaning Extension

Texte intégral

  • 1 Vico (1942), 2.2.1 = §405.

“Quello è degno d’osservazione: che’n tutte le lingue la maggior parte dell’espressioni d’intorno a cose inanimate sono fatte con trasporti del corpo umano e delle sue parti e degli umani sensi e dell’umane passioni. Come ‘capo’, per cima o principio... ‘occhi’ delle viti e quelli che si dicono lumi ingredienti delle case... Lo che tutto va di séguito a quella degnità: che ‘l’uomo ignorante si fa regola dell’universo’”.1

“This is worthy of note: that in all languages the greater part of the expressions concerning inanimate things are formed by metaphor from the human body and its parts and from the human senses and passions. Thus, ‘head’ for top or beginning...; ‘eyes’ for the looped heads of screws and for windows letting light into houses... All this is a consequence of our axiom, that ‘man in his ignorance makes himself the measure of the universe’”.

  • 2 Geeraerts (1997), 6; cf. Geeraerts & Grondelaers (1995), 177: “to a large extent, the synchronic po (...)

1We are now in a position to make some broader statements about the nature and development of the expressions of meaning that we have been considering. While we have observed that this vocabulary can in many cases be used in two different ways – (a) of a speaker or author and (b) of inanimate things, including words and texts – we still need to pad out our theory for how this occurred. One way of approaching this question is to consider how polysemies come into existence – that is to say, the historical background to the double usage of words such as “to mean”. As Dirk Geeraerts puts it, “polysemy is, roughly, the synchronic reflection of diachronic-semantic change”;2 in other words, the double-meanings of a particular word can come about when at a certain point in a word’s history it is applied to a new referent. In this chapter, I argue that a series of such transferences lies at the heart of the double-usage of the classical vocabulary of meaning.

I. Polysemy and Metaphor

  • 3 For some examples, see Bloomfield (1935/1984), 425-495. The philological investigations in diachron (...)
  • 4 The procedure of “calquing” is a form of borrowing that involves translating the components of the (...)
  • 5 Computer operating systems (still) contain visual metaphors, “skeuomorphs”: files look like physica (...)

2Languages accrete for a number of different reasons. One of these involves cultural innovation: when a new concept, technology, or area of inquiry is developed, a vocabulary needs to be created in order to describe it.3 There is no need to look to the distant past in order to find examples of how this occurs – the generation of new items of vocabulary is going on around us today. In the last thirty years we have been faced with the development of a new technology – the Internet – whose apparatus (browsers, search engines, and so on) has required corresponding terminology. Some of these names have resulted from (a) new coinages (neologisms): our ancestors, for example, would not have made much of the verb “to google”, which in the past twenty years has become a common term (strictly speaking, this occurred by denominalization from the name of a company). On the other hand, vocabulary has also been created by means of (b) borrowing it from a foreign language – in German, for example, the English verb “to google” yielded “googeln” and the most popular word for email is “E-Mail” (nowadays often just “Mail”); German speakers did not transfer a term native to their own vocabulary, but instead adopted one from a different language (English).4 A third way of finding new vocabulary, however, is by (c) metaphor and metonymy: in English, we speak of “scrolling down a screen”, “bookmarks”, and “opening files”, just to give three modern instances from the realm of information technology – these expressions were of course already in use with reference to the technology of the book (and related cultural developments) but came to be transferred to a new domain.5

  • 6 On living and dead metaphors, see Searle (1993, originally published in 1979), Traugott (1985), Köv (...)

3Metaphorical transference is certainly a force in language: we do not have an individual, unique word for each and every object, concept, or action in the world, and as it stands the English language possesses a large but clearly finite vocabulary. Metaphor allows us to economize on the number of different words that we use: by means of it we can employ the one word in different ways, transferring it from one application to another, and this is one of the ways that a language’s effective vocabulary increases. Some of these metaphors stick – they become widely used and are eventually incorporated into our standard repertoire, our language; these are sometimes called “dead metaphors” (itself a dead metaphor – metaphors do not literally die), since the novelty of the transference of the word is no longer there.6 Others keep on catching our attention – they remain “alive” and continue to entertain us – and occasionally a new one comes into the world.

  • 7 In other cases, lexicographers may choose to classify the two terms as homonyms and to include the (...)
  • 8 The example is given by Ullmann (1962), 210, and Blank (1999), 62.
  • 9 OLD s.v. “torpeo”: “To be physically numb”, etc; see Varro, De Lingua Latina 5.77.
  • 10 For the distinction between metonymy and metaphor, see Chapter 7.
  • 11 Cf. OED s.v. “bank” n.3, I. 1: “The table or counter of a money-changer or dealer in money”.
  • 12 OED s.v. “bank” n.3, III. 7: “An establishment for the custody of money received from, or on behalf (...)

4When such metaphorical transferences occur, the new usage usually remains classifiable under the same general lemma in the dictionary as the usage from which it issued. In this case the semantic range of the word is broadened (“meaning extension”) and polysemy occurs.7 In English, for example, we can find different sub-entries in the dictionary under the word “torpedo”:8 most of us will associate this word with an underwater propelled explosive charge, but the name was in fact transferred by metaphor from a genus of stingray. The ray’s name is attested already in Varro, who saw it as derived by metonymy from the ray’s ability, presumably because of the sluggishness (“torpedo”) that its shock induced; modern terms for individual species of stingrays include “numbfish” and “crampfish”. The noun “torpedo” in turn was related to the verb “torpere”, “to be sluggish”.9 When the inventor of the new weapon named his creation he evidently thought that the metaphorical transference from stingray to weapon was appropriate. To give an example where metonymy rather than metaphor was the mechanism for the shift,10 we might consider the textbook case of the noun “bank”. Originally, financiers were associated with a counter at which they made transactions (“banca” in Italian);11 when the institution of the bank arose, it retained the original name of “bank”, but came to be included in the dictionary under a different sub-heading.12 The transference from physical object to institution was necessitated by a gap in vocabulary: banks, in the sense of the formal institutions with which we are acquainted, are a relatively recent cultural phenomenon, and language users were forced to find a means of dealing with it.

  • 13 OED s.v. “surf” v, 2.a: “To ride or be carried on the crest of a breaking wave; esp. using a surfbo (...)

5Such transferences can occur in the case of verbs as well as nouns. When we “surf” the Internet we are performing a very different type of action from surfing on a board, but because the two actions are conceptually similar the new usage of the verb has been broadly adopted; once again, there was an opportunity to create an entirely new word in order to cater to the action made possible by a development in technology, but language users opted for a pre-existing term in describing it.13 In fact, the metaphorical transference was so successful that probably nobody would consider creating a fresh piece of vocabulary for the idea of “to visit successively (a series of internet sites)”. In Chapter 7, we shall consider this kind of transference with reference to ancient verbs and expressions of reading – in order to describe the action made possible by the introduction of the alphabet, Greek language users simply adapted older expressions that were conceptually similar to the new action that was in need of a name. Here, however, our point is simply that metaphor and metonymy can expand the range of verbs just as much as that of nouns.

  • 14 OED s.v. “catachresis”: “Improper use of words; application of a term to a thing which it does not (...)
  • 15 Black (1954-1955). Some use the term “catachrestic metaphor”. Cf. Quintilian, Institutio Oratoria 8 (...)

6Another word that we might use to describe this expansion of our vocabulary is “catachresis”. This term has a variety of senses: today, it is often applied to potentially wayward metaphors (“to take arms against a sea of troubles”),14 but I would like to use it in the sense offered by Max Black (following the classical tradition).15 Black defines “catachresis” in the following way:

  • 16 Black (1954-1955), 280-281.

1. “...the use of a word in some new sense in order to remedy a gap in the vocabulary. Catachresis is the putting of new senses into old words”.16

  • 17 As mentioned, it is not the only way; see Bloomfield (1935/1984); Ullmann (1962); Hock & Joseph (19 (...)
  • 18 See the foundational work of Sweetser (1990).

7This is clearly one important way in which languages change and evolve;17 it is in fact the economy afforded by metaphorical and metonymical transference that makes our language a tool of practical value, since if we had a separate word for each individual object we would never be able to master communication. The study of this aspect of language can be a fascinating task, but the difficulty lies in spotting the transferences: to begin with, the new usages of these terms perhaps strike speakers as odd, but it is one of the features of successful catachreses that they become invisible over time. We lose sight of their metaphorical origins – and indeed, the fact that these metaphors and metonymies become formalized as natural and unremarkable parts of our language makes them particularly slippery.18

  • 19 For concise accounts, see, for example, Lakoff (1987), 16-21, and Hanks (2013), 324-329.

8This diachronic aspect to the creation of vocabularies results in a certain “fuzziness” when it comes to category formation and word use.19 The idea that a string of metaphorical transferences may give rise to the synchronic polysemy of a given word explains why it is often difficult to find essences common to all of the things that are associated with the word. J. L. Austin already pointed to this fact in an influential essay (published posthumously):

  • 20 Austin (1970), 72.

2. “Another case is where I call B by the same name as A, because it resembles A, C by the same name because it resembles B, D... and so on. But ultimately A and, say, D do not resemble each other in any recognizable sense at all. This is a very common case: and the dangers are obvious, when we search for something ‘identical’ in all of them!”.20

  • 21 I use Wittgenstein’s example, given at Wittgenstein (1953/2001), §65-71.
  • 22 This point has been taken up by the proponents of prototype theory. See for example Lakoff (1987); (...)

9That is to say, there may be no property essential to all of the things that we call “game”, since the term has been transferred to different types of activities based on resemblances to different members of the earlier family of games.21 This is borne out by a quick consideration of the members of this particular class of activity: one might think that an essential component to all games is that they are amusing, but war games are not designed to be amusing. Another key attribute might be that all games are played by a group of people on different sides, yet games of solitaire are by definition played by only one person. We might think that all games are played for the purpose of winning, but what about a game of ring-a-ring-a-roses? This lack of a unifying element in the case of the word “game” is partly a product of the diachronic development of the word – its metaphorical transference to new practices on the basis of different types of resemblance. In each case, a particular element was key to its inclusion in the broader category of games, but not every member of the category shares the same element.22 The historical explanation for this central issue in the philosophy of language, linguistics, and cognitive science goes back, in part at least, to the role of forces such as metaphor within diachronic language change.

II. Polysemy, Metaphor, and Classical Expressions of Meaning

10This method of retooling words for new purposes is relevant to our discussion of expressions of meaning. The precise contours of the emergence of the polysemies in this vocabulary are difficult to determine, since the evidence that we have from the earliest strata of Greek literature is (a) highly lacunose and (b) depends on the artificial language of epic poetry. There is no way for us to document with any precision the transference of expressions of meaning from animate subjects to inanimate ones (such as texts), since many of the relevant semantic changes occurred at a poorly documented (if not an entirely undocumented) stage. Nevertheless, there are good reasons for believing that such a transference did indeed take place; if the hypothesis is found compelling we shall have uncovered an important metaphorical stratum in the ancient vocabulary of meaning. Let us consider the main expressions that involve an active verb discussed in Chapters 2 and 3 in tabular form:

3. Chapter 2:

Chapter 3:

“(ἐ)θέλειν”

“σηµαίνειν”

“βούλεσθαι”

“µηνύειν”

“νοεῖν”

“φράζειν”

(“ἐννοεῖν”, “διανοεῖν”, “διανοεῖσθαι”)

“ἐµφαίνειν”

“λέγειν”

“δηλοῦν”

“εἶναι”

significare

“δύνασθαι”

ostendere

uelle

declarare

sentire

dicere

sonare

esse

ualere” (“posse”)

(“mǣnan

“to mean”

“to intend”

“vouloir dire”)

  • 23 Ullmann (1962), 214. In the terminology of Lakoff & Johnson (1980), 33-34, this is a species of “on (...)

11If we subtract a few notable examples that have already been flagged (“εἶναι”, “δύνασθαι”, etc.), we can see how much of the vocabulary on the left-hand side of the table relates primarily to mental agency and most notably to the actions of wanting and thinking. As was seen in the previous chapters, these items are often attested of animate subjects (gods, human beings, and animals) before they are found used of inanimate ones (the usage of “νοεῖν” as a verb of S-meaning is only attested from the late fifth century onwards, for example). Moreover, they retained these earlier usages even after they had started to be applied to things such as signs and texts (“νοεῖν” continued to be used of human subjects in the sense of “to notice” in the fourth century). In the face of the evidence that we have to hand, it is more plausible to argue that they were transferred from animate to inanimate objects rather than the other way around. It is in fact very common to find such “anthropocentric” transferences; the phenomenon has been remarked upon for some time (as may be noted from the quotation from Vico with which I opened this chapter) and continues to be studied by cognitive linguists under the rubric of “embodiment”.23

12When we consider the items discussed in Chapter 3 (listed in the right-hand column on the table above), we are dealing with verbs of a different nature: as we have seen, terms such as “σηµαίνειν” and “significare” do not refer to mental activity as such but rather to the act of showing or sign-giving. Nevertheless, the argument for metaphorical transference is strong here as well: in the case of several of these verbs the usage of human beings is either attested earlier than that of inanimate objects or can be understood as being prior on the basis of the word-form itself. The verb “σηµαίνειν”, for example, is only used in epic poetry of animate subjects in the act of sign-giving, ordering, and making a mark, and in fact the first applications of the verb to writing are to be found in the fifth century (e. g. Herodotus, Histories 5.35.3). The history of the verb “significare” apparently mirrors this process, the very structure of the verb (“to make a sign”) suggesting that it has been transferred: from an etymological perspective, when we say “the sign signifies” we are actually using a somewhat circular expression. The ancient languages were the basis for our own, and we have inherited the metaphorical language contained within them – even though their metaphorical quality is no longer readily apparent because the history and components of “to signify” are no longer commonly understood.

13We must nevertheless account for the usage of expressions such as “εἶναι”, “δύνασθαι”, “esse”, “ualere”, and “posse”, which do not fit our paradigm: as mentioned in Chapter 2, one does not say “Cicero est” for “Cicero means/intends”, and from this it is clear that not all of the classical expressions of meaning issued from the vocabulary for human meaning, wanting, speaking, and sign-giving. The verbs “εἶναι” and “esse” would in fact appear to be among the most basic conceivable expressions of meaning, in that they simply establish equivalence between two nouns in the nominative by serving as a copula; in English we can also ask the following:

4. “What is ‘army’ in French?”.

  • 24 One might compare the Sanskrit “śakti” (generally, “power”, “ability”, “strength”), as employed in (...)
  • 25 See also my earlier comments on the relationship between words and coinage (Chapter 2, footnote 101 (...)

14Verbs of the “εἶναι”/“esse” type are, it should be noted, somewhat limited in what they can do as expressions of meaning; they are often used of equivalences between things (including lexical items), but are less useful for commenting on more sophisticated types of meaning: one cannot, for example, use the phrase “the story is X” in the same way as “the story means X”. On the other hand, when it comes to the verbs “δύνασθαι”, “ualere”, and “posse” (in the sense of “to mean/signify”), we might suggest that a different metaphorical transference from the one that we have been considering is in play – the relevant meaning came about through an extension of the earlier sense of “power” or “ability”. Although the usage of “δύνασθαι” of words is not attested in Homer, it is quite prominent from Herodotus onwards, the term conjoining two items that are literally “able to do” the same thing, or have the same “force” or “valence”.24 The terminology might therefore be assimilated, to a certain degree, to the other expressions of meaning, in that they are rooted in a conception of agency – in the word’s ability to accomplish something.25

  • 26 See, for example, Fantham (1989); Hutchinson (2013). For verbs of meaning, one should also consider (...)
  • 27 On this, see recently Bettini (2012); McElduff (2013). For an extended representation of Roman pseu (...)
  • 28 “...Graecos interpretabantur” (“they clarified the meaning [?] of Greek authors” Suetonius, De Gram (...)
  • 29 Stilo composed works on etymological lore and wrote a commentary on the Carmen Saliare, activities (...)
  • 30 For a list of Latin terminology for translation, see McElduff (2013), 189-196 (appendix).

15Finally, we should say a word about the influence of the Greek expressions on the Latin ones. Latin literature and criticism came into being as a reflex to and in accordance with the developed Greek models,26 and it is plausible to hypothesize that this relationship is reflected in the formation of the Latin vocabulary of meaning. The similarities between expressions such as “δύνασθαι” and “ualere” suggest that the latter is a semantic calque created in order to translate a Greek term of art, and we might posit a similar influence on the development of “significare” by “σηµαίνειν”. Indeed, Latin literature was in its earliest phases very much a project of translation;27 this would seem to suggest that a vocabulary of meaning existed early on in Roman literary history, a hypothesis that finds strength in the fact that Terence was already demonstrably using “sibi uelle” in the transferred sense in the first half of the second century BC. In fact, Suetonius claims that Livius Andronicus and Ennius were already performing interpretations of Greek authors from the end of the third century BC onwards (although the evidence for this remains somewhat uncertain).28 It is possible that the Roman vocabulary of meaning underwent an expansion with the burgeoning study of poetry, rhetoric, and grammar at Rome in the third and second centuries BC, potentially through the agency of grammarians such as Aelius Stilo and his predecessors.29 In this regard it should be pointed out that the Latin vocabulary of translation was itself the result of metaphorical transference: that is, the language’s verbs for “to translate” are based on earlier vocabulary (“uertere”, “transferre”, etc.).30 Whatever the case, the existence of the Greek vocabulary of meaning meant that the Latin equivalents did not need to be invented entirely anew: the presence of expressions for what things mean in Greek would in fact have given rise to similar locutions in all of the languages whose speakers were influenced by the tradition of Greek culture. Given this, the creation of the Greek vocabulary would appear to be of primary importance, and we shall concentrate on its early history in the following chapter.

III. The Role of Economy

  • 31 See Grice (1989) on Conversational Maxims, and Sperber & Wilson (1995), 46-49. One well-known examp (...)

16We have already discussed (and shall return to) one possible motivation for the transference of this terminology – the development of a new technology or area of inquiry. Here, however, I would like to discuss a further consideration that would have contributed to the acceptance of the transference. Linguists have long emphasized the role of economy in how we communicate with each other – given the investment of time and energy involved, we tend, at least, to express ourselves with as much brevity as possible. When we do not observe the principle of economy, our listener may in fact infer that something ulterior is being implied.31 This has a particular relevance to the series of transferences that we are considering, since one way of making communication more economical (ceteris paribus) is to cut down on the number of words we use in order to express a given thought. Cicero already noted that metaphorical usages can aid economy within language:

5. nonnumquam etiam breuitas translatione conficitur, ut illud ‘si telum manu fugit’. imprudentia teli missi breuius propriis uerbis exponi non potuit, quam est uno significata translato.

“Often brevity is achieved by metaphorical transference, for instance ‘if the weapon fled the hand’: it was not possible to express the unintentional nature of the dropped weapon more concisely by means of the proper terms than it is conveyed by a single transferred word” (Cicero, De Oratore 3.158).

17Here, a verb that is typically used of sentient beings (“fugere”) is employed in a transferred sense of an inanimate object – a weapon. The sense of the utterance is communicated most economically by using a term in a metaphorical fashion. The employment of transferred expressions of meaning can increase the brevity of a passage along similar lines: when it is unnecessary to specify who is doing the “meaning” (this information being implicit), it is possible in English simply to leave out the agent behind the message – the context of the utterance determines the correct interpretation. Thus, rather than using an expression such as the following:

6. “The individual who made the statement meant X”.

18one may simply say:

7. “The statement meant X...”.

19Both of these phrases may refer to the meaning of the statement-maker, but number 6 strikes us as awkward, perhaps over-explicit, and number 7 takes a briefer time to say; in accordance with this, this latter formulation will often be preferred in communication, and 6 will only be employed if 7 is perceived to be ambiguous.

20The relevant syntactic structures of Greek and Latin would have exerted similar pressure on speakers: it simply takes a longer time to refer explicitly to the intentional agent as well as the mode of communication, and this is certainly true of the verbs that we discussed in Chapters 1, 2, and 3 (“νοεῖν”, “διανοεῖν”, “βούλεσθαι”, and so on). At times, this would have given language users an incentive to apply verbs of meaning to inanimate objects and symbols rather than to the intentional agents who manipulated them: for instance, rather than make explicit mention of the role of Histiaeus, who had the slave’s head tattooed in order that he might send a message to Aristagoras, Herodotus could phrase things in the following way:

8. τὰ δὲ στίγµατα ἐσήµαινε, ὡς καὶ πρότερόν µοι εἴρηται, ἀπόστασιν.

“The tattoos signified (as I have already mentioned) revolt” (Herodotus, Histories 5.35.3).

21The metaphorical transference would thus have been partly motivated by the simple practicalities of communication.

22Similar phenomena would also have been active in the case of the relevant nouns: in English, one can say both “the meaning of the individual who made the statement” or “the meaning of the statement”, but the second option is more economical. As was suggested earlier, the transference of nouns appears to have occurred via either (a) ellipsis, whereby the speaker was simply left implicit in the syntax, or, more likely, (b) metonymy, whereby the intention or thought of the speaker came to be attached to the meaning of linguistic expressions themselves. In both cases, however, we are dealing with a form of shortening (brachylogy). While it is jarring to see the words “νόος” and “uoluntas” employed of things that do not have a mind or will of their own, we might note that this type of articulation has the advantage of economy outlined above; take, for instance, the following use of “uoluntas”:

9. in finitione quae sit uoluntas nominis quaeritur.

“In definition, it is inquired after the meaning of a term” (Quintilian, Institutio Oratoria 7.10.1).

23In the case of number 9, we can see how it is unnecessary to explicitly refer to any speaker; the meaning of the term can stand on its own without reference to an actual language user. With the advantage of economy came, of course, a certain ambiguity, as it is sometimes difficult to determine whether the meaning of the utterer or the meaning of the sentence as a compositional unit is a tissue. In most cases context will decide, but there are situations in which the distinction is less than clear-cut; this has important implications for modern criticism, which will come under discussion in Chapter 9.

IV. Polysemy, Metaphor, and Modern Expressions of Meaning

24When we turn to the modern languages in light of the previous discussions, certain issues begin to resolve themselves. It is in fact its explanatory power in this regard that makes the argument for metaphorical transference compelling. It explains, for example, the polysemy in our contemporary lexicons under the entries for the term “meaning” and the fact that we can use this expression in such divergent ways as the following:

10. “I meant that that milk was in the fridge”.
“I
meant to hit the ball”.
“Beer
means everything to me”.
“What do wars
mean for humanity?”.
“Your note did not
mean anything”.

25We can here discern the fossils of a series of old transferences whose history we can no longer trace: their antiquity is shown by the presence of the polysemy of the relevant terms in Anglo-Saxon and Old High German. Similar lists, with various additions and subtractions, may be replicated for the Romance languages in the case of the expressions “vouloir dire”, “volere dire”, “querer decir”, and so on; since the polysemy of these Romance expressions does not appear to be reflected in early Latin we might assign the genesis of the idiom to the period of late antiquity when the vernaculars were beginning to separate into their modern forms. This variety of applications has certain theoretical implications – and we shall discuss these further in later chapters – but here it will be worth mentioning what this view entails for our prospects of ordering and correcting our usage of language.

26First, the polysemy that eventuated from these metaphorical transferences has resulted in the fact that it is impossible for the modern philosopher, linguist, or philologist to find a single, comprehensive definition for the modern English noun “meaning” and its verb “to mean”. Just as we saw in the case of the word “game”, there is no specific feature that is common to all types of meaning: the word has become, from the lexical point of view, irreducible to an axiom. This is important, because it is quite common to read strong views of what “meaning” can and cannot be – for some, “meaning is use”, for others “all meaning is realized at the point of reception”, for others “the meaning is the author’s own”. When considered in the light of the previous surveys, however, it is clear that each of these aphorisms is in fact false when taken in its strictest sense. Each makes a claim about what meaning is, but each ignores the other usages of the relevant term. The fact is that this particular chess piece can perform different moves.

27The history of the polysemy that we have been considering goes against the “optimistic” view that our language is completely rational, built up of distinct blocks that, if used properly, will be free from all ambiguity. It would (in some ways) be desirable if we could define the term “meaning” in a single fashion – restricting it to the intentions of human beings, for example: only entities with minds could, on this model, mean things, by virtue of the fact that only minds can intend things. It would thus be possible to regulate language and prevent misapprehensions of what a word like “to mean” is doing – there would be complete consensus on the ways in which it could be employed. Our modern vocabulary of meaning would be split according to the following table:

11. “intend”

“signify”

“to mean”

“to signify”

“meinen”

“bedeuten”

“vouloir dire”

“signifier”

28The first group (“intend”) characterizes the meanings of individuals when they intend to do something: it makes use of the language of mental life (thinking, wishing, and so on), and would therefore seem to be the domain of sentient beings. The second (“signify”) seems to provide us with a vocabulary for talking about the meanings of texts to us (their significance).

29As we have seen, however, the division does not hold up in practice; a number of verbs of the first category (“to intend”) can be shown to serve the same function as those that belong to the second (“to signify”) – they “move across”, as it were, to be applied to things. Even apparently unambiguous verbs such as “to intend” itself may be, and frequently are, transferred to texts in modern prose – just as Plato used the expression “τί σοι ἄλλο νοεῖ τοῦτο τὸ ῥῆµα” and Gellius “uocabulis idem sentientibus” (see the Introduction and Chapter 2). On the other hand, it is not the case that verbs from the second group can come to play the same role as the first; the English verb “to signify”, for example, cannot be employed as a synonym for “to intend”. It is impossible to substitute the phrase “I signified X” for “I intended X”. We therefore see a basic asymmetry in the relationship between these two categories: the items on the left-hand side of the table above can “move across”, but this movement is not reciprocal (if the sense of “meaning” is to be preserved).

30Still, the fact is that the items on the right-hand side of the table (“to signify”) can also be used of human beings. People signify things, both in the sense of “revealing” and “making a sign” to each other, and the word is certainly not used only for what inanimate things mean. This has a bearing on the overall argument concerning the metaphor of text = person; even when we use the term “to signify” as opposed to “to mean” in sentences such as “the text signifies”, we do nothing in particular to inhibit the possible interpretation of the text as an agent – especially when the term is used in conjunction with other verbs that connote intentionality. This dovetails with the general observation put forward in the Introduction – that every active verb that we use in English in order to describe what texts do can also be used of human beings. Any attempt to prescriptively regulate these expressions of meaning is unlikely to get off the ground; it turns out that our vocabulary for what texts do is fundamentally metaphorical and there is no way that the transferences can be undone. While we may wish to control the vocabulary at our disposal for talking about what texts do, the attempt to retroactively purify our language of earlier associations is impossible – we must work with what we have.

  • 32 On the impulse towards the rationalization of language, see Eco (1995).

31Of course, it is questionable as to whether the removal of this ambiguity from our language is in fact something to be desired; while the creation of languages in which no ambiguity is present has been a frequent aspiration in human history, the grammars of the resultant tongues have typically been cumbersome and only a very limited few have been able to converse in them.32 The syntactic and semantic ambiguities that thread through the natural languages are in fact highly productive; on the one hand, it is hard to envisage an author such as Baudelaire coming to prominence without them, but on the other they also pervade our everyday discourse and serve a number of practical functions. Far from revealing a speaker as confused and imprecise, the exploitation of these fissures marks one out as skilled in speech – rhetoric could not exist without the imprecision and latitude for expression afforded by the natural languages, particularly when it comes to the tropes of metaphor and metonymy. Language, and human life itself, is all the richer for its ambiguity, so long as it does not confound us.

V. Conclusion

32In this chapter we have put forward a thesis for how the polysemy found in a number of key expressions of meaning (both ancient and modern) came into being on the basis of the classical examples collected in Chapters 2 and 3. I have argued that a number of expressions of meaning used of inanimate objects (including texts) developed out of vocabulary for mental activity, and primarily from that of volition and intention. In addition, the double usage of verbs and nouns of meaning was partly motivated by the principle of economy. A number of questions, however, still remain: is there any evidence for the history of this transference in the literary record (beyond the instances already noted)? How was the metaphorical quality of the Greek and Roman expressions of meaning “activated” in literature itself? Were there any related phenomena that lent support to the personification of the text? And what of the metaphors for “metaphor” itself?

Notes

1 Vico (1942), 2.2.1 = §405.

2 Geeraerts (1997), 6; cf. Geeraerts & Grondelaers (1995), 177: “to a large extent, the synchronic polysemy of lexical items is a reflection of their diachronic development”. For an introduction to certain aspects of semantics (particularly prototype theory) with reference to ancient Greek, see Clarke (2010).

3 For some examples, see Bloomfield (1935/1984), 425-495. The philological investigations in diachronic semantics from the late nineteenth and early twentieth centuries (by figures such as Hecht, Paul, and Breal), which did much to elucidate the role of metaphor and metonymy in linguistic change, bear notable similarities to the cognitive approach of the past decades; on this, see Geeraerts (2010), 277.

4 The procedure of “calquing” is a form of borrowing that involves translating the components of the source word; in Anglo-Saxon, for example, the Latin term “proscribere” (“to proscribe”) was transformed into “forewrītan” (BT s.v. “forewrītan”: “To proscribe, banish”). To give a classical example, Cicero (De Finibus 1.17) calques the Democritean word “ἄτοµοι” as “corpora indiuidua” (“indivisible bodies”); cf. TLL s.v. “indiuiduus”. Lucretius (or an earlier Roman author), in his effort to communicate the theory of Epicurean physics without recourse to Greek, rendered earlier atomist terms such as “ἀρχαί”, “σώµατα”, and “σπέρµατα” in Latin as “principia”, “corpora”, and “semina”. On these, see Bailey (1947), 2.606-608. Lucretius refers on multiple occasions to what he terms the “patrii sermonis egestas” (“the poverty of our native language” De Rerum Natura 1.832, 3.260, cf. 1.136-139). On this theme, see Fögen (2000), particularly 61-76.

5 Computer operating systems (still) contain visual metaphors, “skeuomorphs”: files look like physical files, and the icon of the trashcan at the bottom-right hand corner of the screen represents the function of deleting a file.

6 On living and dead metaphors, see Searle (1993, originally published in 1979), Traugott (1985), Kövecses (2002). See also Chapter 8 for further considerations of metaphors surrounding “metaphor”.

7 In other cases, lexicographers may choose to classify the two terms as homonyms and to include the derivative in the dictionary as a separate entry.

8 The example is given by Ullmann (1962), 210, and Blank (1999), 62.

9 OLD s.v. “torpeo”: “To be physically numb”, etc; see Varro, De Lingua Latina 5.77.

10 For the distinction between metonymy and metaphor, see Chapter 7.

11 Cf. OED s.v. “bank” n.3, I. 1: “The table or counter of a money-changer or dealer in money”.

12 OED s.v. “bank” n.3, III. 7: “An establishment for the custody of money received from, or on behalf of, its customers”. For a fascinating discussion of the difficulties of dealing with the polysemy of the word “bank”, see Hanks (2013), 66-69.

13 OED s.v. “surf” v, 2.a: “To ride or be carried on the crest of a breaking wave; esp. using a surfboard”; 4.a: “trans. To visit successively (a series of internet sites)...”.

14 OED s.v. “catachresis”: “Improper use of words; application of a term to a thing which it does not properly denote; abuse or perversion of a trope or metaphor”.

15 Black (1954-1955). Some use the term “catachrestic metaphor”. Cf. Quintilian, Institutio Oratoria 8.6.34, who translates the term as “abusio”; “eo magis necessaria κατάχρησις, quam recte dicimus ‘abusionem’, quae non habentibus nomen suum accommodat, quod in proximo est” (“on account of this, κατάχρησις, which we rightly call ‘abusio’, is all the more important; it adapts the nearest available term to describe something for which no actual term exists...”). Further comments on the word “catachresis” can be found in Chapter 8.

16 Black (1954-1955), 280-281.

17 As mentioned, it is not the only way; see Bloomfield (1935/1984); Ullmann (1962); Hock & Joseph (1996), 215-317; Geeraerts (1997), 84-122; Blank (1999); Fortson (2003); and Hanks (2013), 145-172.

18 See the foundational work of Sweetser (1990).

19 For concise accounts, see, for example, Lakoff (1987), 16-21, and Hanks (2013), 324-329.

20 Austin (1970), 72.

21 I use Wittgenstein’s example, given at Wittgenstein (1953/2001), §65-71.

22 This point has been taken up by the proponents of prototype theory. See for example Lakoff (1987); Geeraerts (1997); see also Geeraerts (2010), with references to the early work of Eleanor Rosch.

23 Ullmann (1962), 214. In the terminology of Lakoff & Johnson (1980), 33-34, this is a species of “ontological metaphor”. For a survey of relatively recent work on embodiment, see Evans & Green (2006).

24 One might compare the Sanskrit “śakti” (generally, “power”, “ability”, “strength”), as employed in writing on grammar and interpretation; cf. MW s.v. “śakti”: “the power or signification of a word”.

25 See also my earlier comments on the relationship between words and coinage (Chapter 2, footnote 101).

26 See, for example, Fantham (1989); Hutchinson (2013). For verbs of meaning, one should also consider the language of early divination; cf. Cicero, De Diuinatione 1.93.

27 On this, see recently Bettini (2012); McElduff (2013). For an extended representation of Roman pseudo-translation in action, see Plautus, Poenulus 930-1031.

28 “...Graecos interpretabantur” (“they clarified the meaning [?] of Greek authors” Suetonius, De Grammaticis et Rhetoribus 1.2). See Kaster (1995), 52-54, who stresses the need for caution: “It is at least questionable that Suet. knew very much about the actual practices of Livius and Ennius. It is rather more likely that he was projecting the practices of his own times... on to the scrappy data he had gathered about the earliest period” (53). Ennius discussed issues of translation: Musas quas memorant nosce nos esse Camenas (“those whom they call ‘Muses’, know that we are the Camenae” Ennius, Annales fr. 487 Skutsch). For a fragment from Ennius’ Medea involving interpretation of words, cf. “fructus uerborum aures aucupant” (“[my?] ears catch the harvest of the words” Ennius, Medea fr. 245 Jocelyn).

29 Stilo composed works on etymological lore and wrote a commentary on the Carmen Saliare, activities that would presumably have required a sophisticated vocabulary of meaning beyond the simple “sibi uelle”. On Aelius Stilo, see Kaster (1995), 68-70.

30 For a list of Latin terminology for translation, see McElduff (2013), 189-196 (appendix).

31 See Grice (1989) on Conversational Maxims, and Sperber & Wilson (1995), 46-49. One well-known example is the addition of the irrelevant piece of information “John is very punctual” in a recommendation letter for an academic position; the violation of the maxims of efficiency and relevance means that the hiring committee can assume that it is being tipped off to the fact that there is something wrong with the candidate.

32 On the impulse towards the rationalization of language, see Eco (1995).

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Acheter

Volume papier

amazon.fr
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search