por Jan Harald Alnes, University of TromsØ
0 Some books are interdisciplinary in the true sense of the word, that is to say, they treat a field of study that overlap between different disciplines without simply, or more or less, applying the insights from one field onto another field. This book (henceforth called “False Friends”), published in the Series Routledge Studies in Linguistics, is such a book. As the title says, the theme of the book is false friends, which, roughly, but to give the reader some ideas, could be explained as follows: It frequently happens that two ordinary languages, called “the Source language” (SL) and “the Target language” (TL), have homographic or homophonic words or phrases in common. If the meanings of two such words or phrases are either unrelated, or partly, but not completely, overlap, we have a case of false friends. Obviously, some of the false friends between SL and TL are false friends by more or less coincidence, while others are etymologically related. The former are called “chance false friends” and the latter “semantic false friends.” Of course, there are a number of subclasses under these headings. The most interesting subclass, and accordingly the one that receives most attention in this book, is that of partial semantic false friends. To exemplify, the Spanish “inexcusable” and the English “inexcusable” are partial semantic false friends as the former might be translated by the two English terms “inexcusable” and “unavoidable,” depending on the context (False Friends: 7 –8). Chamizo-Domínguez presents an enormous sample of such cases, each one of them exploring various quite interesting facts about the development of language, both internally and between different, more or less related, languages. The notion of context, just alluded to, turns out to be the key to understanding the present book. I will discuss it in some detail in due time.
Chamizo-Domínguez explores the general phenomenon of false friends from a “linguistic” as well as a “philosophical” point of view. His overall focus is Translation theory, a huge and growing interdisciplinary subject of which Spain is the leading centre. The present book is, beyond doubt, a major contribution to one of the main themes of this field of study. Translation between languages, or in contexts, that have, or contain false friends is approached from a theoretical, or “semantic” perspective, as well as from a practical, or “pragmatic” perspective. Let me note that as a philosopher, I am somewhat foreign to the use of real experiments in order to test out various assumptions and/or underscore points; I am more used to what philosophers call “thought-experiments.” I really enjoyed the different tests that Chamizo-Domínguez undertook towards students—some new to translation theory, others experienced—, as well as towards colleagues. It is of particular interest to note that it frequently happened that the test-persons provided translations that surprised or at least was unexpected from the author’s point of view (False Friends: 43 –45 and 159 –164). Such tests are used to illuminate both semantic and pragmatic issues.
This book, then, could, and should, be reviewed from quite different angles. Another interdisciplinary researcher, a linguist, or a philosopher may review it. As a philosopher by profession, I have my own limited approach towards, and insight into, this field of research. It is accordingly my hope that some member of either of the two other groups also would review this fine piece of work. (Fortunately, it has come to my attention that such is indeed the case!)
The review consists of five parts. I begin by introducing the general ontological theme in the philosophy of language, viz. the question about the nature of meaning. Some philosophers claim that meanings are entities of some kind or another, while others argue that meaning should be taken care of in some other way, prominently by way of an account of use. I detect both tendencies in False Friends. This question therefore reoccurs in different clothing throughout the four subsequent Parts. The major aim of this review is to read False Friends consequently as falling within the second kind of approach towards meaning. Consequently, I find it necessary from time to time to rephrase or reconstruct the actual wording of False Friends, while attempting to keep Chamizo-Domínguez’s insights. But, of course, as will become evident in my discussions, I also have other aims. Part 2 focuses on the notion of synonymy; this highly troublesome term is a key-term in False Friends, and, according to its author, in Translation theory quite generally. In part 3, I discuss what the author calls “secondary meaning,” in particular with respect to metaphors. Donald Davidson’s theory of meaning and metaphor is invoked in order to throw a critical light on this complicated issue. The notion of an implicature is crucial to False Friends, and a central issue in Part 4 is accordingly Paul H. Grice’s theory of implicatures. This theory, in turn, is but a central feature of Grice’s general and unified theory of meaning and communication. I present some aspects of this theory and relate them to False Friends. Since Chamizo-Domínguez in certain aspects, but not in all, is close to Grice, I believe that the latter’s theory is particularly useful in order to throw illuminating light on the central topics of False Friends. The review is rounded off in Part 5. I make a few comments on the pragmatic aspects of the book, and provide a few observations about the relationship between Willard van O. Quine and Chamizo-Domínguez. It should be noted from the outset that the various issues treated below are interconnected in a number of ways. Thus, I have found some overlap unavoidable.
2 Chamizo-Domínguez takes the notion of synonymy to be the central notion in his project:
From these five concepts [synonymy, homonymy, polysemy, register, and diachrony], one is both the most basic and based for the others and for any theory on translation: this is synonymy, which has also been the most strongly criticised one by linguists and philosophers of language, though (False Friends: 32).
Hence, should we have to summarize to the minimum what the task of translation involves, I cannot think of anything better than understanding translation as identifying synonyms in two different languages. If this intuitive opinion is right, then the notion of synonymy in itself and its analysis become cornerstones in order to perform any attempt for establishing a theory of translation (False Friends: 33).
…despite all this, the introduction to the notion of synonymy in this work is essential in order to identify the cases in which two given terms in two given languages may be interchanged in the translation of an utterance (False Friends: 36).
In the first and third quote, Chamizo-Domínguez alludes to the by now classical philosophical and linguistic objections to the notion of synonymy; he subscribes to Quine’s objections in particular (Quine 1953,1960). Chamizo-Domínguez solves his problem in the only way possible, viz. by introducing a definition of synonymy that both avoids the actual philosophical and linguistic issues and is adequate for his purposes:
I shall deal with a weak notion of synonymy … two terms would be synonymous if, in a given context, one can be substituted by the other with no changes in the truth values of the utterance in which the substitution takes place (False Friends: 40, italics added).
The three key features of this definition is firstly, that synonymy is defined for terms, and not, say, sentences, secondly, the invocation of the term “context”, and finally, that the substitution is to preserve, not the truth-condition of the sentence or utterance, but its actual truth-value. In order to spell out the significance of this definition, I treat these features separately.
(i) It is clearly correct to give the definition of synonymy with respect to subsentential expressions, whether it is to terms only, as is Chamizo-Domínguez choice,[iv] or also is to include phrases and expressions. (This particular theme cannot, due to limit of space, be pursued any further, but the reader should be aware that it is a quite interesting and important issue). By the way, Quine observes that this is in accordance with the etymology of “synonymous” (Quine 1960: 61). It therefore comes as a surprise that Chamizo-Domínguez immediately after providing his definition, maintains that it is on line with the corresponding definition provided by Keith Allan in his comprehensive textbook in linguistics, viz. “A is synonymous with B only if when A is true, then B is true, and vice versa. (It follows that if A is false then B is false, and vice versa)” (False Friends: 40, Allan 2001: 115). Here synonymy is defined as a relation between sentences or utterances; it is, after all, sentences or utterances, and not terms, which have truth-values. Allan’s definition is clearly useless, as sentences substantially unrelated might, by coincidence, always have the same actual truth-value. Allan is aware of this, and introduces later on a modal version of his definition (Allan 2001: 188). To me, it is evident that even this refined definition of synonymy fails for a number of reasons. And, as a matter of fact, Chamizo-Domínguez is very close to committing himself to my view when he introduces four strong definitions of synonymy, where one of them is based on the notion of necessity, which he finds useless (False Friends: 31 –33). Now, since these definitions take terms as their starting point, Allan’s definition is not included among them. Still, it is hard to believe that one could accept the uselessness of the four strong definitions, and retain Allan’s definition. To conclude this discussion, synonymy ought to be defined for terms (or other subsentential expressions) in the first place, only then might one attempt to define the derived notion of synonymy of sentences or utterances. And this is what Chamizo-Domínguez does.
(ii) Let us turn our attention to the term “context.” A characteristic feature of one of the two important classes of partial semantic false friends (the other important class is mentioned in a moment), is the fact that in one setting, a biological one, say, they can be substituted for each other while preserving truth-value (or truth-conditions, see below), while in another setting, for instance a juridical one, they cannot so be substituted. (If the two terms could be substituted in all such contexts, they would be what we might call “completely synonymous”). Chamizo-Domínguez introduces a number of such contexts in his book, for instance, ecclesiastical contexts (False Friends: 39), scientific contexts, philosophical contexts, linguistic contexts (False Friends: 124), legal or political contexts (False Friends: 25) and academic contexts (False Friends: 126). I find this manner of singling out a subclass of partial semantic false friends by the invocation of contexts illuminating and clear. The notion of a context is not defined, but it might be viewed as a generalization of the old logical notion of a domain of discourse, or a subject of discourse. It seems, however, as if the word “context” is used in another, more committing, sense when Chamizo-Domínguez maintains that the sentence, or metaphor, “You are the cream in my coffee” is a context in which the term “cream” is used (False Friends: 48). We will discuss this particular example in some detail in Parts 3 and 4.
Now, a word about the other main subclass of partial semantic false friends, namely terms that, despite being completely synonymous or synonymous in given contexts, still ought not to be translated for each other. This lack of translatability is due to what is often called “pragmatic factors.” The reason is that the two terms have different “register.” For some reason or other, the author does not define the term “register,” but his examples clarifies what it is. Let us take a brief look at one of them. Note first that “hemorroide” and “almorrana” are synonyms in Spanish, and can be translated into the English synonyms “haemorrhoid” and “pile,” respectively. Says Chamizo-Domínguez:
…It is obvious that any Spanish native speaker will find their register different. That is particularly why, although we may interchange “hemorroide and “almorrana” and keep the principle of substitution salva veritate, the implicatures of the sentences involved will be very different. Thus, the use of “hemorroide” often implies a certain cultural level of the speaker, while for “almorrana” the speaker’s cultural level is supposed to be lower (False Friends: 53).
He further points out that if a given language has only one word for haemorrhoids, then that word will be a partial semantic false friend to the corresponding two words in Spanish and English, due to a lack of corresponding implicatures. (Implicatures are discussed at some length in Part 4, we leave it undefined for now.) Chamizo-Domínguez concludes his discussion by saying that “although the reference of all those terms is the same, the implicatures will vary depending on the case” (False Friends: 53).
Clearly, then, the notion of partial semantic false friends that depends solely on register is a finely grained notion, much finer than e.g. truth-conditions; see below. Of course, although the term “register” might be quite new, the phenomenon has been known for a while. Gottlob Frege gives numerous corresponding examples in his attempts at clarifying his notion of a thought. I shall close this discussion with one of his remarks that also has some bearing on our next issue:
… in this connection it is useful to the poet to have at his disposal a number of different words that can be substituted for one another without altering the thought, but which can act in different ways on the feelings and imagination of the hearer. We may think e.g. of the words “walk”, “stroll”, “saunter” [“gehen”, “schreiten” and “wandeln”]. These means are also used to the same end in everyday language. If we compare the sentences “This dog [Hund] howled the whole night” and “This cur [Köter] howled the whole night”, we find that the thought is the same. The first sentence tells us neither more nor less than does the second. But whilst the word “dog” is neutral as between having pleasant or unpleasant associations, the word “cur” certainly has unpleasant rather than pleasant associations and puts us rather in mind of a dog with a somewhat unkempt appearance (Frege 1979: 140). [v]
(iii) We have reached the most complicated and demanding part of our discussion of Chamizo-Domínguez’s definition of synonymy, namely his invocation of the principle of substitution salva veritate. This substitution principle was first formulated by Leibniz: “Eadem sunt, quae sibi mutuo substitui possunt, salva veritate,” or alternatively “Eadem sunt, quorom unum potest substitui alteri salva veritate.”[vi] John L. Austin translates the principle, in the latest phrasing, as follows: “Things are the same as each other, of which one can be substituted for the other without loss of truth” (Frege: 1980: 76). [vii] It is common to take the principle to relate to the notion of identity, eventually to provide a definition or explication of that notion. Frege is famous for having developed an intricate philosophy of mathematics and language on the basis of the fundamental insight that two singular terms might refer to the same object, while having different sense. Such is the case with the two Greek terms “Hesperus” (“The Morning Star”) and “Phosphorus” (“The Evening Star”). These two terms might be substituted for each other salva veritate (but not in opaque contexts): “Hesperus is the brightest star seen in the east before sunrise” and “Phosphorus is the brightest star seen in the east before sunrise” have the same true value. But, says Frege, since “Hesperus” and “Phosphorus” have different sense, the thoughts expressed by the two sentences differ.[viii] Chamizo-Domínguez, on the other hand, and in strong opposition to Frege, relies on the salva veritate principle when defining synonymy. Thus, as he himself states, his notion of synonymy is weak, we could call it “referential synonymy.” It states than any two members of a pair of co-referring or co-extensional terms, in a given context, are synonymous. Maybe, I wonder, it is too weak. Let me elaborate a bit. Despite deep divergences in general philosophical outlook, such disparate philosophers as Frege, Wittgenstein, Carnap, Quine and Davidson, just to mention the most outstanding ones, maintain that the meaning (“sense,” in the case of Frege) of a sentence is given when one has stated the conditions under which it is true.[ix] Accordingly, then, a truth-conditional semantics might contain this rephrased version of Chamizo-Domínguez’s definition:
Two terms would be synonymous if, in a given context, one can be substituted by the other with no changes in the truth-conditions of the utterance in which the substitution takes place.
Let “TV” be shorthand for the original definition, and “TC” shorthand for this new one. Now, let me underscore that both TV and TC take for granted that there is a deep connection between meaning and truth, and it is hard to imagine an account of meaning that does not. The crucial difference between them is that the while TV involves the actual truth-value, TC avoids invoking reference or extension. Let us return to Frege’s example once more. According to TV, “The Morning Star” and “The Evening Star” are synonymous in quite a number of contexts, while they are in general not synonymous according to TC. Clearly, then, the definition of synonymy in terms of truth-conditions is more finely grained that the definition of synonymy in terms of truth-value. That is to say, all pairs of terms that are synonymous according to TC, are synonymous according to TV, but not vice versa. It is highly important to note that the distinction between partial semantic false friends due to lack of synonymy in a given context and partial semantic false friends due to different registers in a given context, holds for both definitions. Recall here that Frege himself time and again underscored the difference between what we have called “implicatures” and truth-conditions. (The substitution of “cur” for “dog” preserves truth-conditions, but invokes implicatures. In these cases, the implicatures are conventional and not conversational; but see Part 4 for a further discussion and modification of this claim.) I have, in other words, not been able to figure out the exact reasons for defining synonymy as coarse-grained as done in False Friends. To rephrase my present point as a challenge: Although “synonymy” in these contexts is a quasi-technical terms, is it not the definition in term of truth-conditions, rather than the definition in terms of truth-value, that is in accordance with our vernacular language? Now, it is highly important to note that Chamizo-Domínguez has made a strong case for the claim that whether one opts for TV or TC, or another “weak” definition of synonymy, the restriction to context should be a crucial part of the definition. This is an improvement over traditional weak definitions.
Let me close of this discussion of synonymy by repeating my agreements with Chamizo-Domínguez. First of all, I fully agree with him both that synonymy must be given a weak definition, i.e. the definition should not involve modality, and that the definition must be given with respect to a particular interest. (The interest-relativity of philosophical notions is, by the way, one of the major lessons to be learned from the writings of Quine.) I further find it crucial with respect to Translation theory to draw a theoretical distinction between synonymy and register.
3 Our next theme is levels of meanings, in particular metaphorical meaning. I am not in favour of Chamizo-Domínguez’s liberate use of the notion of meaning, in particular when he maintains that a word has primary, secondary and transitional meanings. There seems to be too many meanings around. I have, of course, no objections against this way of speaking, per se, as long as it is made clear that it does not carry ontological commitments. We are moving into muddy waters, and it might well be that my worry is due to such terminological matters; but the issue needs to be explored.
(i) Let us begin by taking yet another at the notion of a context and its role in the definition of synonymy. I shall propose an alternative analysis of Chamizo-Domínguez’s problematic example; an analysis I believe to fit perfectly into the main line of reasoning in False Friends. Chamizo-Domínguez notes that the English word “doctor” and the Spanish word “doctor” are partial semantic false friends; in some contexts, i.e. academic ones, they might be substituted for one another, but not in general in medical contexts, this since one distinguishes in Spanish between “doctor” and “médico” in cases where one simply uses “doctor” in English (False Friends: 120 –123. I grossly simplifies Chamizo-Domínguez detailed and intriguing treatment of this case). Let us, for the sake of illustration, say that the terms are partial, but not completely, synonymous. In the medical context the two terms cannot always be substituted salva veritate. Compare this to “You are the cream in my coffee,” where the sentence is said to be the context, and it is maintained that the English word “cream” and the Spanish word “crema” are synonymous in this context. But, one could ask, what does this mean? What does it mean to say that “You are the cream in my coffee” and “Eres la crema de mi café” have the same truth-value? In other words, what sense are we to make out of the claim that “You are the cream of my coffee” has any truth-value at all? To me, the sentence (or an utterance of it in any regular contexts, with given utterer and audience) is patently false. Let us apply the definition of synonymy to this case. As the context is just one sentence, it follows that any word that, when substituted for “cream” in “You are the cream of my coffee” makes it false, are synonymous with “cream;” an unwelcome consequence, indeed. To be fair, let me quote Chamizo-Domínguez’s own words:
When a metaphor is proposed for the first time (a novel metaphor) in a linguistic system, speakers understand the use of such terms as a diversion from its literal meaning—a “flouting” or a “categorical falsity”, as Grice called it (1989: 34). In this case the metaphor seems to be a usage matter, and its interpretation is occasional and limited to the moment of the utterance. Thus, the meaning of a novel metaphor would not go beyond its particular sense at the very moment of the utterance. Therefore, the example of metaphor that Grice himself provides, … “You are the cream of my coffee” … could be understood in Spanish without many difficulties if literally translated as …“Eres la crema de mi café.” This is because the English noun “cream” and the Spanish noun “crema” share the meaning of “the oily or butyraceous part of the milk, which gathers on the top when milk is left undisturbed” (OED) and “sustancia grasa contenida en la leche”, or “nata de leche” (DRAE), in English and Spanish, respectively. Consequently, in a context such as [“You are the cream of my coffee”], “cream” may be replaced by “crema” with no changes in the truth values of the sentences involved (False Friends: 48).[x]
Although I do not think this account works, I believe that the solution to our problem is simple. Why not simply maintain that with respect to the translation of a metaphor, the synonymy principle (whether TV or TC) is invoked secondary in the following manner: As “cream” and “crema” are synonymous in regular standard (standard, normal, most) contexts, “crema” is the foremost candidate for a translation of the actual sentence into Spanish? An account along this line would not only solve our problem, it would in addition rely on a clear, unified and simple notion of context. To me, this proposal is but a minor modification of Chamizo-Domínguez’s own theory, and I see no reason why it should not be acceptable to him. One might note that if two terms are partial semantic friends, and this is due to contexts (in the sense of “subjects of discourse”) which are common, then the problem of translating the terms in novel metaphors is similar to the problem of literal translation; also a consequence welcomed by Chamizo-Domínguez, I presume. My present suggestion about translation furthermore fits nicely into what I am saying about Grice’s account of literal meaning in Part 4.
(ii) Chamizo-Domínguez’s treatment of Donald Davidson’s influential theory of metaphors is somewhat hostile (False Friends: 47ff). The reason, in my opinion, is mainly that Davidson does not clarify his notions of meaning and use, nor the distinction between them, adequately for readers who are not intimately familiar with his philosophy and the tradition to which he belongs. One needs, in particular, always to be strictly aware of the fact that Davidson uses the term “meaning” in a quasi-technical sense. Or more accurately, he uses “meaning” with a specific task, or interest in mind, to recall Quine’s methodological insight. That use, although schematically spelled out elsewhere, is simply taken for granted when Davidson presents his theory of metaphor. In fact, the difference between Davidson’s account of metaphors and that of the author of False Friendsmight be slighter than the latter thinks. I begin by providing a (very) schematic account of Davidson’s theory of meaning (drawn from Davidson 1984: Part 1); thereafter I compare his theory of metaphors to the one provided in False Friends.
Davidson’s grand project in the philosophy of language (I am here ignoring his attempts at formulating a unified theory of language and action) is to develop a theoretical framework for a satisfying theory of meaning for natural languages. Such a theory, he maintains, must be a theory about the truth-conditions for the actual as well as the potential sentences of that language. He argues at length and in detail, that a Tarskian definition of truth is such a theory. (The principal difference between Tarski’s own historical project and that of Davidson, is that while Tarski defines truth and takes meaning for granted, Davidson starts out with a primitive notion of truth and uses it to provide a theory of meaning.) Such a theory must be recursive and it must contain specifications of the reference of the singular terms as well as a satisfaction-relation for the predicates. Thus we get the so-called T-sentences of the form
“P” is true inL if and only if S,
where “`P´” is a sentence in the language “L” and “S” is a canonical description of “`P´” in the actual meta-language. A theory of meaning is adequate and extensionally correct, according to Davidson, just in case it generates all and only the correct T-sentences for the actual language. To illustrate, let Spanish be the object-language and English the meta-language. Then this is an example of a T-sentence:
“La nieve es blanca” is trueS if and only if snow is white.
We note, then, that in this theory of meaning, we do not need any notion of meaning. Let us instantiate our standard example of a metaphor, “You are the cream in my coffee” (assume for the same of the argument that the reference of “Eres” has been fixed). We get
“Eres la crema de mi café” is trueS if and only if you are the cream in my coffee.
Thus, “Eres la crema de mi café” is simply false. Such is the case with most metaphors (Davidson 1984: 257). The fallout is that the significance of metaphors cannot be captured by way of Davidson’s theory of meaning; in fact, he even makes the stronger claim that no theoryof meaning can make sense of metaphors. The adequacy of a theory of meaning is limited to the literal meaning of words. This is explicitly maintained in the introduction to (Davidson 1984):
No discussion of theories of meaning can fail to take account of the limits of application of such theories. The scope must be broad enough to provide an insight into how language can serve our endless purposes, but restricted enough to be amenable to serious systematization …`What Metaphors Mean´, is mainly devoted to the thesis that we explain what words in metaphor do only by supposing they have the same meanings they do in non-figurative contexts. We lose our ability to account for metaphor, as well as rule out all hope of responsible theory, if we posit metaphorical meanings (Davidson 1984: xix).
What then, from the perspective of this approach towards meaning, is the significance of metaphors? First of all, note that metaphors are an important and probably unavoidable feature of all complex use of language, including the scientific one:
In the past those who have denied that metaphor has a cognitive content in addition to the literal have often been out to show that metaphor is confusing, merely emotive, unsuited to serious, scientific, or philosophical discourse. My views should not be associated with this tradition. Metaphor is a legitimate device not only in literature but in science, philosophy and law; it is effective in praise and abuse, prayer and promotion, description and prescription (Davidson 1984: 246).
This is followed up a bit later:
We must give up the idea that a metaphor carries a message, that it has a content or meaning (except, of course, its literal meaning) … No doubt metaphors often make us notice aspects of things we did not note before; no doubt they bring surprising analogies and similarities to our attention; they do provide a kind of lens or lattice, as Black says, through which we view the relevant phenomena … What I deny is that metaphor does its work by having a special meaning, a specific cognitive content (Davidson 1984: 261 –262).
As is clear from these passages, within the framework of a theory of meaning, the only acceptable notion of content, or cognitive content, with respect to sentences or utterances is the one captured by the T-sentences: they provide the meaning of the sentences of the language, period. But, metaphors are none the less highly important—an opinion clearly expressed in the quote above.
Hopefully, my sketch has provided at least an idea of Davidson’s view on metaphors. Let us turn to the criticism that Chamizo-Domínguez voices against Davidson’s account:
Davidson’s denial to talk about metaphorical meanings … departs from the real or methodological oversight of how signifiers attain new meanings, but do not necessarily lose the former one, and how this process takes place in practice. The hardcore of this oversight resides in not distinguishing the three stages a metaphor … may undergo; e.g. novel, semilexicalised and lexicalised (False Friends: 47).
Here, I shall not discuss the significance of a novel metaphor; that issue is simply too huge to be covered in these pages. (For the record, note that Davidson, as one might expect, attaches no importance at all as to whether a metaphor is novel or not (Davidson 1984: 252f).) Rather, I shall make an attempt at establishing that Davidson, despite Chamizo-Domínguez’s claim to the contrary, do have the resources needed in order to distinguish between the three stages a metaphor may undergo. Let us take a look at Chamizo-Domínguez’s main case against Davidson:
… the most interesting stage of metaphors is semilexicalisation because speakers are aware of both the literal and the figurative meaning of a certain term. These metaphors allow us to establish conceptual nets and conform a system to conceptualise a particular reality in terms of another different reality … Let us look at a quote from the daily press to illustrate this point:
 “La Plaza de las Ventas, primera en el mundo y cátedra del toreo, aguarda nueva gestión” [Las Ventas bullring, top of the world and bullfighter’s chair, is awaiting for new management … El País, December 1 2002, p. 41).
If, after having read it, we look up the entry for “cátedra” at the DRAE, we will see that none of the nine senses and five collocations of this word makes reference to bullfighting at all. Hence in  “cátedra” does not mean “aula” [classroom], empleo y ejercicio del catedrático [chairman´s position], or “facultad o material que enseña un catedrático” [subject thought by a chairman] (DRAE), for instance. Consequently, in , we are providing a new transferred meaning for “cátedra” as a synonym of “lugar en que se practica lo mejor del toreo” [place where the best bullfighting is played or “lugar en que se puede aprender lo mejor de la tauromaquia” [place where one can learn the best in bullfighting matters]. But when using “cátedra” as in , we are actually conceptualizing bullfighting in terms of academy. Then,  is coherent with a wide conceptual network where we use terms with academic literal referents to denote and conceptualize the taurine domain (False Friends: 49).
This obviously needs to be spelled out. When Chamizo-Domínguez provides his two alternative synonyms for “cátedra,” he presupposes, since the synonyms do not involve any reference to, or do not even allude to, academy, that they in the actual context differ from “cátedra” exclusively in terms of register. Thus, the two alternative substitutions preserve truth-value, but lose the important implicature (in our wide sense of the term, see Part 4). It seems clear then, that Chamizo-Domínguez takes the truth-value of  to depend on whether La Plaza de las Ventas in fact has the quality of being the place where the best bullfighting is taking place, or alternatively, of being the place where one can learn the best in bullfighting matters, while Davidson simply takes  to be false. (But as a metaphor, it might still be effective and reach its intended effects.) It is not clear to me what our intuitions say about this case; it is not even clear what is at stake. This point is reinforced by looking at the further part of Chamizo-Domínguez criticism. For, when he talks about “conceptualising,” this might be read strongly or weakly. According to the strong reading, the conceptualization in terms of academy contains sentences (utterances/thoughts/propositions) that are true or false; according to the weak reading, we think about or encounter bullfighting along academic lines, but this does not involve sentences (utterances/thoughts/propositions) that are true or false, rather it is a manner of approaching bullfighting. Now, clearly, Davidson has no problems at all subscribing to the weak reading, but he would surely object to the strong reading, which seems to be that of Chamizo-Domínguez. But, again, how are we to decide between these two alternative ways of understanding the intended effects of the chosen metaphor? My general point is that despite Chamizo-Domínguez impressive phenomenological descriptions of the content and significance of metaphors, Davidson has the resources to transform these descriptions into his own theoretical framework. This exemplifies what one might call “effective redescription.” (It is furthermore evident that Davidson might take account of the practical example that Chamizo-Domínguez discusses (False Friends: 50), by way of his own resources.) To conclude: it seems to me that the difference between Davidson and Chamizo-Domínguez boils down to as to whether the latter takes metaphors not only to be appropriate, adequate, and illuminating, but in addition to have truth-values. If he does, he takes such terms such as “secondary meaning” and “linguistic meaning” to be non-reducible, but then he owes us an account of the truth-makers that justifies ascribing these metaphors another truth-value than the literal one.
4 Chamizo-Domínguez refers to Grice’s notion of implicatures several times. Recall that Grice talks about different kinds of implicatures, the conventional and the conversational ones, in particular. [xi] Roughly, the general distinction is that while a conventional implicature is generated by way of a word or expression, independently of knowledge of the involved context, a conversational implicature must be calculated by way of some extra-linguistic knowledge.[xii] As Grice’s focus is mainly on the conversational implicatures, he mentions, to my knowledge, only two examples of conventional implicatures. This is the first one:
In some cases the conventional meaning of the words used will determine what is implicated, besides helping to determine what is said. If I say (smugly), He is an Englishman; he is, therefore, brave, I have certainly committed myself, by virtue of the meaning of my words, to its being the case that his being brave is a consequence of (follows from) his being an Englishman. But while I have said that he is an Englishman, and said that he is brave, I do not want to say that I have said (in the favored sense) that it follows from his being an Englishman that he is brave, though I have certainly indicated, and so implicated, that this is so I do not want to say that my utterance of this sentence would be, strictly speaking, false should the consequence in question fail to hold. So some implicatures are conventional …” (Grice 1989: 25f).
The crucial thing to be noted here is that what is said is closely connected to—it might even be considered an Ersatz of—the traditional notion of a proposition or a Fregean thought; it is the part of the significance of the utterance which, strictly speaking, is true or false. Or, to speak alternatively, what is said is spelled out by spelling out the truth-conditions of the utterance. Since it is of such an extremely importance to grasp this notion, both in order to understand Grice, and in order to get hold of my discussion below, I will cite a long telling passage from Grice:
In the sense in which I am using the word say, I intend what someone has said to be closely related to the conventional meaning of the words (the sentence) he has uttered. Suppose someone to have uttered the sentence He is in the grip of a vice. Given a knowledge of the English language, but no knowledge of the circumstances of the utterance, one would know something about what the speaker had said, on the assumption that he was speaking standard English, and speaking literally. One would know that he had said, about some particular male person or animal x, that at the time of the utterance /whatever that was), either (1) xwas unable to rid himself of a certain kind of bad character trait or (2) some part of x´s person was caught in a certain kind of tool or instrument … But for a full identification of what the speaker had said, one would need to know (a) the identity of x, (b) the time of the utterance, and (c) the meaning, on the particular occasion of utterance, of the phrase in the grip of a vice … this brief indication of my use of say leaves it open whether a man who says (today) Harold Wilson is a great man and another who says (also today) The British Prime Minister is a great man would, if each knew that the two singular terms had the same reference, have said the same thing (Grice 1989: 25). [xiii]
This determination of what is said plays a role both when I discuss Grice’s second example of a conventional implicature, and then, afterwards, when I present Grice’s account of literal or conventional meaning.
Let us look at Grice’s second example of a conventional implicature, used by Frege already in 1879 (Frege 1972: 123). An utterer (“U”) has uttered the sentence “She was poor but she was honest.” “What U meant, and what the sentence means,” says Grice, “will both contain something contributed by the word “but,” and I do not want this contribution to appear in an account of what (in my favored sense) U said (but rather as a conventional implicature)” (Grice 1989: 88). To spell out, U says the same by an utterance of “She was poor and she honest” as he says by an utterance of “She was poor but she was honest,” this since both utterances have the same truth-conditions: they are true if and only if the actual woman is poor and she is honest. Now, obviously, U´s choice of using “but” rather than “and” indicates that he thinks there is a certain contrast between the property of being poor and the property of being honest. Thus U applies the linguistic fact that the use of “but,” in opposition to “and,” in general indicates a contrast between the content of the two conjuncts. Since this contrast is built into the very word, the implicatures generated by its use are conventional. Now, when Chamizo-Domínguez maintains that uses of synonymous words with different registers generate implicatures, he must be thinking of conventional implicatures. A true follower of Grice must furthermore maintain that it is by making a choice between using “haemorrhoids” or “piles”, say, that U generates an implicature. A crucial feature of Grice’s theory is that the generation of an implicature per se is intentional (cf. the passage cited in note xiv). Thus, if U knows the word “pile,” but is unfamiliar with “haemorrhoids,” then, in Grice’s sense, even though U’s use of the word “pile” tells something about him, the use does not generate an implicature. That is to say, that which the actual use tells about its user is not part of what is communicated, or part of the significance of the utterance. As Chamizo-Domínguez does not raise the issue about intentionality, it is not clear to me that this is what Chamizo-Domínguez has in mind. But, in any case, I am not sure that Grice would even label the discussed intentional kind of indication “an implicature,” as both his examples involves logical considerations, while the “implicature” under consideration is meant to tell, indirectly, something about the utterer. It might well be then, that Chamizo-Domínguez uses the term “implicature” in a looser sense than Grice. On the face of it, I see no problem with this, as Chamizo-Domínguez’s project differs from that of Grice. Still, it would be an interesting project, and certainly one to the taste of Chamizo-Domínguez, to figure out whether any theoretical insight is gained by splitting the overarching class of (conventional) implicatures, in the loose sense of the term, into different subclasses.
In the long passage cited above, Grice uses the terms “literal meaning” and “conventional meaning.” And, as he makes abundantly clear, the generation of a conversational implicature, or some other kind of non-conventional implicatur, depends on this meaning. Now, Grice treats metaphors as conversational implicatures. That is to say, one understands the literal meaning, that is, what is said (in the present technical sense of the term “said”), plus, in certain complex cases, what is conventionally implicated, and then one works out, or calculates, the intended message.[xiv] This means that a metaphor derives from the literal meaning of the involved words. But now, what is conventional or literal meaning to Grice? The answer to this question, I believe, might indicate a path to a further development of the theory of metaphors that is to the taste of Chamizo-Domínguez. I believe this since the answer ties together Grice’s theory of meaning and his theory of implicatures. It is only possible, of course, to give a rough sketch of this complex account here.[xv] As is well known, Grice analyses meaning in general as constituted by complex sets of higher-order intentions. Now, obviously, the simpler a complex of intentions is, the better is the changes of getting the message across; and this is the clue to Grice’s understanding of the notion of literal meaning. The following passage is stripped off any of the usual technicality (variables, quasi-variables, etc.) in Grice’s discussions. I quote it at length, since it contains a number of points closely connected to those of Chamizo-Domínguez:
The general suggestion would therefore be that to say what a word means in a language is to say what it is in general optimal for speakers of that language to do with that word, or what use they are to make of it; what particular intentions on particular occasions it is proper for them to have, or optimal for them to have. Of course, there is no suggestion that they always have to have those intentions: it would merely be optimal, ceteris paribus, for them to have them. As regards what is optimal in any particular kind of case, there would have to be a cash value, an account of why it is optimal. There might be a whole range of different accounts. For example, it might be that it is conventional to use this word in this way; it might be that it is conventional among some privileged class to use it this way—what some technical term in biology means is not a matter for the general public, but for biologists; it might be, when an invented language is involved, that it is what is laid down by its inventor. However, what we get in every case, as a unification of all these accounts, is the optimality or propriety of a certain form of behaviour (Grice 1989: 299).[xvi]
Let me illustrate with a simple example. Suppose I want to convey the view to someone that I find the staff at the department of philosophy at the University of Málaga generous and helpful. Depending on the situation and the audience, there might be different ways of conveying this message. This we know from Grice’s deep analyses into the conditions that govern conversations. But it is just one, or a few ways, that would not involve calculations or, more generally, would not depend on any particular knowledge or presuppositions among the utterer and the audience, and that is to utter the sentence: “I find the staff at the department of philosophy at the University of Málaga generous and helpful.” As this sentence is the optimal way of communicating the intended message, it is the conventional or literal way of communicating it. (In this connection I do not take account of the complicating factors that involve the distinction between sentence-meaning and word-meaning.) Now, note that “and” and “but” have different literal meanings, as they are used to convey different messages (Grice 1989: 121). Literal meaning, in other words, goes beyond what is said. My present suggestion is that the phenomenon of partial semantic false friends might in general be explained as the fact that a particular word has some optimal use in one language that it does not have in another language. Furthermore, according to the present account, the distinction between the two synonyms “haemorrhoids” and “piles” is explicable as different optimal uses systematically tied to social class or status in the English-speaking population. Lexicalization, to speak with Chamizo-Domínguez, happens when a word gains an optimal use that it did not use to have. The phenomenon of semi-lexicalization that is so important to the author of False Friends, might also be taken care of from this theoretical perspective. We would then talk about a non-conventionalimplicature (in the wide sense of the word) that normally or generally are generated by the use of a given word. (Two words are partial semantic false friends if, despite being synonyms, the use of one of them would generate an implicature of this kind, while the corresponding use of the other would not.) Such an implicature corresponds nicely to Grice’s “generalized conversational implicature”—the kind of implicatures that really interested Grice (Grice 1989: 37–40).
To sum up the admittedly rather sketchy discussion of this Part, I have, as in the former Part, attempted to take account of Chamizo-Domínguez’s insights, while avoiding troublesome non-reducible notions of secondary meanings, transitional meanings, metaphorical meaning, and the like. This made me to reformulate some of his basic notions into a Gricean or quasi-Gricean vocabulary—a vocabulary, by the way, already partly utilized in False Friends. But to repeat my warning once again, it might well be that my misgivings in this Part are due to terminological matters, I am not sure.
5 Except from a minor observation in a note, I have not talked about the pragmatics of False Friends, as this theme is beyond my competence. I found the last chapter of False Friends fascinating, and I learned a lot from it. It was especially fun to see that the main practical example was a translation of a text on existence by Quine—a major theme of this review! Before closing off, I shall briefly mention a further point about Quine, having to do with the distinction between the theoretical and the practical, a distinction that to a certain degree overlaps with Chamizo-Domínguez’s distinction between semantics and pragmatics. Chamizo-Domínguez uses strong formulations when he underscores the role of a notion of synonymy for Translation theory:
The practical effect of expelling the notion of synonymy from the linguistic theory will not be anything but denying the mere possibility of translation, or at least its determination, and, consequently, giving up any attempts to establish an axiological criterion that allows us to choose between the different possible translations of a text or an utterance. Or, as the lesser of two evils, accepting the possibility of the existence of different alternative manuals to translate the same text, all of them compatible with the data recorded from native speakers and all of them incompatible among them, as Willard van O. Quine claimed: “Two translators might develop among them independent manuals of translation, both of them compatible with all speech behavior, and yet one manual would offer translations that the other would reject. My position was that either manual could be useful, but as to which was right and which was wrong there was no fact of the matter” (False Friends: 36).
Now, since Chamizo-Domínguez accepts Quine’s objections to any general and philosophically illuminating notion of synonymy, he also accepts the “lesser of the two evils.” In fact, due to the long history of cultural interchange among the various cultures in the world, as well as the established translations between a huge numbers of languages, this “evil” has little, if any, “practical effect.” (In general, even hitherto unknown cultures or languages are related to some known cultures or languages, and thus the translation of such languages will not be radical in Quine’s sense.) Quine’s point is purely theoretical: it is directed against prevalent theories of meaning and conceptual content. We could, if we were ingenious enough, even translate the humdrum words of our fellow speaker of English (“chair,” “man,” “Spanish”) non-automatic and non-homophonic, “while” to quote Quine, “conforming to all his dispositions to verbal responses to all possible stimulations.” After this observation, Quine sets out the philosophical consequence of his thought-experiment:
Thinking in terms of radical interpretation of exotic languages has helped make factors vivid, but the main lesson to be derived concerns the empirical slack in our own beliefs. For our own views could be revised into those attributed to the compatriot in the impractical joke imagined; no conflicts with experience could ever supervene, except such as would attend our present sensible views as well. To the same degree that the radical translation of sentences is underdetermined by the totality of disposition to verbal behavior, our own theories and beliefs in general are under-determined by the totality of possible sensory evidence time without end (Quine 1960: 78).
Chamizo-Domínguez adds flavour to Quine’s vision of language by demonstrating, by way of a real experiment, that even when translating between related languages of related cultures, the choice between different translation-manuals has real effects. Quine would most certainly appreciate the closing passage of False Friends chapter 5:
And the problem of a variety of possible manuals for the translation of a single text seems to lie, in this case, mainly in the fact that the possibilities of expression of one language do not coincide with those of another. In these cases, although the translator may know that the manual he chooses has flaws, and that other manuals exist (which also contain flaws), he has to choose one of them, perhaps the one he considers least faulty. Taking all of that into account, most of translations we can provide shall reveal some aspects of the original text and hide many others as well because, as Quine himself asserts “The radical translator is bound to impose about as much as he discovers”… What Quine says about the radical translator, can be said, mutatis mutandis, about the normal translator, as shown throughout this book, and in particular in this chapter (False Friends: 164).
Let me close off by noting that a reader must be intimately familiar with the great European languages not to learn something from Chamizo-Domínguez’s almost encyclopaedic knowledge, expressed in examples spread out on as good as every page. The author utilizes the web and the dictionaries to be found there in an exemplary manner. This is an illuminating and fascinating book, which I would have liked to be longer than its 186 pages. Hopefully, Chamizo-Domínguez at some occasion or another responds to some of my challenges, as I am convinced that this would lead to fun reading, with a number of well-funded and illustrative—always in more than one way—examples, backing up his theoretical points. *
Allan, Keith (2001) Natural Language Semantics, Oxford: Blackwell.
Alnes, Jan Harald (1999) “Sense and Basic Law V in Frege´s Logicism,” Nordic Journal of Philosophical Logic 4, 1 –30.
Chamizo-Domínguez, Pedro, J. (1987) “La traducción como problema en Wittgenstein,” Pensamiento, 43/170, 179 –196.
Chamizo-Domínguez, Pedro, J. (1998) Metáfora y conocimiento, Málaga: Analecta Malacitana
Chamizo-Domínguez, Pedro, J. and Igor E. Klyukanov (in progress) “E. Bakthin and J. Ortega y Gasset: Nostradad and Beyond.”
Davidson, Donald (1984), Inquiries into Truth and Interpretation, Oxford: Clarendon Press.
Frege, Gottlob (1972) Conceptual Notation, in Terrell W. Bynum (ed. and trans.) Frege: Conceptual Notation and related articles, Oxford: Oxford University Press, pp. 103 –204.
Frege, Gottlob (1979) Posthumous Writings,Hans Hermes, Friedrich Kambartel, Friedrich Kaulbach (eds.), Peter Long, Roger White (trans.) Oxford: Basil Blackwell.
Frege, Gottlob (1980) The Foundations of Arithmetic, John. L. Austin (trans.), Evanston, Illinois: Northwestern University Press.
Frege, Gottlob (1988) “On Sense and Meaning,” in Peter Geach, and Max Black (eds.), M. Black (trans.) Translations from the Philosophical Writings of Gottlob Frege, 56 –79.
Grice, Paul H. (1989) Studies in the Ways of Words, Cambridge, Mass.: Harvard University Press.
Ishuguro, Hidé (1990) Leibniz´s Philosophy of Logic and Language, Cambridge: Cambridge University Press.
Mates, Benson (1986) The Philosophy of Leibniz: Metaphysics and Language, New York: Cambridge University Press.
Neale, Stephen (1992) “Paul Grice and the Philosophy of Language,” Linguistics and Philosophy, 15, 509 –599.
Platts, Mark de Bretton (1979) Ways of Meaning, London: Routledge & Kegan Paul.
Quine, Willard v. O. (1953) “Two Dogmas of Empiricism”, in Quine, From a Logical point of View, Cambridge, Mass.: Harvard university Press.
Quine, Willard. v. O. (1960) Word & Object, Cambridge, Mass.: The M.I.T. Press.
Jan Harald Alnes,
University of TromsØ
[i] Indirectly in this work, but explicitly in Chamizo-Domínguez (1987) and in Chamizo-Domínguez and Klyukanov (in progress).
[iii]Of course, I do not think anything I am saying here would persuade a firm believer in propositions to give up her view. There are no knockdown arguments in philosophy, to speak with Robert Nozick.
[iv]I am not sure that he always sticks to this restriction, however. At a certain point it is maintained that a term is synonymous to an expression (False Friends: 49). I discuss this particular case in Part 3.
[v]The quote is from an unpublished article called “Logic,” and the telling title of the chapter is “Separating a Thought from its Trappings” (“Trennung des Gedankens von den Umhüllungen”). This article is well worth studying for a reader interested in the present subject.
[vi]The first is from (Frege 1988: 64), the second from (Frege 1980: 76).
[vii]Of course, the exact understanding of this principle is a subject of some rather heated discussions, cf. e.g. (Mates 1986) and (Ishuguro 1990) for different readings of the principle.
[viii]This treatment is simplified in a number of ways. I discuss Frege’s understanding of sense and identity in detail in (Alnes 1999).
[ix]See the references in (Platts 1979: 3).
[x]“DRAE” is shorthand for: “Diccionario de la lengua española” and “OED” is shorthand for “Oxford English Dictionary.”
[xi]Grice suggests that there are other kinds of non-conventional implicatures than the conversational ones, but he does not substantiate this possibility (Grice 1989: 26 and 28).
[xii]In Grice´s own words:
“I wish to make [a distinction] within the total signification of a remark: a distinction between what the speaker has said … , and what he has implicated …, taking into account the fact that what he has implicated may be either conventionally implicated (implicated by virtue of the meaning of some word or phrase which he has used) or nonconventionally implicated (in which case the specification of the implicature falls outside the specification of the conventional meaning of the words used)” (Grice 1989: 118).
The crucial notion of “said” is specified below.
[xiii]Note that this notion of said is closely related to Davidson’s notion of said in (Davidson 1984: 246).
[xiv]The following long passage is informative both about Grice’s way of distinguishing conventional and conversational implicatures, and about his way of specifying the content of a conversational implicature. Note also that it is clear that an implicature must be intended:
“The presence of a conversational implicature must be capable of being worked out; for even if it can in fact be intuitively grasped, unless the intuition is replaceable by an argument, the implicature (if present at all) will not count as a conversational implicature; it will be a conventional implicature. To work out that a particular conversational implicature is present, the hearer will reply on the following data: (1) the conventional meaning of the words used, together with the identity of any references that may be involved; (2) the Cooperative Principle and its maxims; (3) the context, linguistic or otherwise, of the utterance; (4) other items of background knowledge; and (5) the fact (or supposed fact) that all relevant items falling under the previous headings are avaliable to both participants and both participants know or assume this to be the case. A general pattern for the working out of a conversational implicature might be given as follows: ”He has said that p; there is no reason to suppose that he is not observing the maxims, or at least the Cooperative Principle; he could not be doing so unless he thought that q; he knows (and knows that I know that he knows) that I can see that the supposition that he thinks q is required; he has done nothing to stop me thinking that q; he intends me to think, or is at least willing to allow me to think, that q; and so he has implicated that q” (Grice 1989: 31).
The reader should compare this passage to Chamizo-Domínguez’s pragmatic account of how to work out a message in (False Friends: 146 –151). To me at least, that treatment suggests that Chamizo-Domínguez’s understanding of communication and the task of translation is close to that of Grice.
[xv]Stephen Neale underscores the significance of taking Grice to have a unified theory of meaning and conversation in his influential (Neale 1992). My sketch of Grice’s theory is deeply inspired by this article. It must in particular be noted that the way I connect the notions of optimality and literal or conventional meaning is worked out in detail by Neale.
[xvi]The more technical treatment of this idea, but withoutthe significant evaluative notion of optimality is given in (Grice 1989: 126 –37). In a rudimentary form, the idea is present already in Grice’s classical article ”Meaning” from 1957 (Grice 1989: 220 –223).
* Pedro Chamizo-Domínguez has informed me in conversation that he holds the view that metaphors have truth-values as metaphors, and that his argument is to be found in (Chamizo-Domínguez 1998: 71 –94). Thus, I pinpointed correctly the difference between him and Davidson in Part 3, but in light of this information there is, despite my attempt at arguing the contrary, a deep substantial difference between the two authors. I have not had the opportunity to study Chamizo-Domínguez’s book on metaphors, and must leave an assessment of it for another occasion. Instead, I will sketch a reading of Grice that suggests that, at least to a certain degree, he is in agreement with Chamizo-Domínguez. In the long passage cited in note xiv above, Grice noted that the “presence of a conversational implicature must be capable of being worked out,” and a bit later he gave this scheme for calculating such an implicature:
“He has said that p; there is no reason to suppose that he is not observing the maxims, or at least the Cooperative Principle; he could not be doing so unless he thought that q; he knows (and knows that I know that he knows) that I can see that the supposition that he thinks q is required; he has done nothing to stop me thinking that q; he intends me to think, or is at least willing to allow me to think, that q; and so he has implicated that q.”
If we focus on such phrases as “he has said that p,” “he thought that q,” “he thinks that q,” and “he has implicated that q,” we realize that the variables “p” and “q” take as their values something which is said, or can be said, in the Gricean sense. (Let us ignore, as it is at present irrelevant, the issues concerning conventional implicatures.) It follows that since Grice takes metaphors to be conversational implicatures, they have truth- values. In this sense, then, Chamizo-Domínguez and Grice agree against Davidson. But, on the other hand, while Chamizo-Domínguez maintains that metaphors have truth-values asmetaphors (whatever that exactly means), Grice argues that when a conversational implicature is present, the audience has replaced it, or its wordings, by something said, or something that can be said. And it is this that is true or false. Recall here, that one and the same wording, “You are the cream in my coffee,” might be calculated differently, leading to different “that q”-results, depending on the utterer and the audience (see e.g. the two different results in (Grice 1989: 34). I will not speculate any further on these issues at this moment, only note that I need to undertake further investigations in order to figure out the exact relationship between the respective accounts of Grice and Chamizo-Domínguez on metaphors.