Varieties of Tone

Frege, Dummett and the Shades of Meaning

R. Kortum

• Publisher: Palgrave Macmillan
• Year: 2013
• ISBN:
  9781137263544
• Topic: Philosophy
• Subtopic: Language in Philosophy

In clear and lively prose that avoidsjargon, the author carefully and systematically examines the many kinds of subtly nuanced words or word-pairs of everyday discourse such as 'and'-'but', 'before'-'ere', 'Chinese'-'Chink', and'sweat'-'perspiration', that have proven resistant to truth-conditional explanations of meaning.

Part I

Historical Preliminaries

1.1

Historical Preliminaries

The business of raising and pursuing answers to questions concerning language or concerning the meanings of particular expressions goes way back. It does not originate with Frege. Ideas and themes central to the philosophy of language surface frequently in the dialogues of Plato, for instance. The main topic in his Cratylus, written nearly 2,500 years ago, concerns the question of names (nominal terms) and whether they are affixed to their bearers on a natural or a conventional basis that can be judged to be true or false, correct or incorrect. Aristotle’s approach to the metaphysics of Being is to seek a definition of the verb ‘to be’. Plato’s renowned student comes up with no less than five distinct senses of the term. I doubt whether even former U.S. President Bill Clinton appreciated just how much depends on the meaning of ‘is’. Aristotle, who first worked out a system of formal deductive relations, devoted considerable attention to matters of language and meaning. The opening section of his treatise On Interpretation runs like this:

First we must define the terms ‘noun’ and ‘verb’, then the terms ‘denial’ and ‘affirmation’, then ‘proposition’ and ‘sentence’. Spoken words are the symbols of mental experience and written words are the symbols of spoken words. Just as all men have not the same writing, so all men have not the same speech sounds, but the mental experiences, which these directly symbolize, are the same for all, as also are those things of which our experiences are the images.

... As there are in the mind thoughts which do not involve truth or falsity, and also those which must be either true or false, so it is in speech. For truth and falsity imply combination and separation. Nouns and verbs, provided nothing is added, are like thoughts without combination or separation; ‘man’ and ‘white’, as isolated terms, are not yet either true or false. In proof of this, consider the word ‘goat-stag’. It has significance, but there is no truth or falsity about it, unless ‘is’ or ‘is not’ is added, either in the present or in some other tense (2006: 28).

Here we see a genuine philosophical concern with language, with the difference between word-meaning and sentence-meaning, and with the appropriate application of the predicates ‘is true’ and ‘is false’. Grappling with Aristotle’s syllogistics, medieval philosophers and theologians distinguished primary from secondary words in a sentence. Primary words, or categoremata, are nouns and verbs; together these form complete sentences. Secondary, or ‘syncategorematic’, terms do not signify anything by themselves, but are meaningful only in conjunction with the subject and predicate of a sentence. Adjectives, adverbs, conjunctions, prepositions, and the like were of special interest to logicians at the time insofar as some of them, like the word ‘not’, impact the truth and falsity of sentences in which they occur. One of the key debates concerned the status, vis-à-vis this distinction, of the verb ‘is’: is it truly part of a predicate expression in such sentences as ‘Socrates is mortal’, or does it function merely as an inert ‘copula’, an essentially meaningless connector of subject and predicate?

During the great Scholastic debates over the primacy of existence between universals and particulars, the 12th-century’s greatest and most original thinker, the nominalist Peter Abelard, anticipated the modern distinction between sense and reference. In his Logica ‘ingredientibus’ (ca. 1115–1119) Abelard drew a distinction between two semantic properties belonging to the broad class of nominal expressions: these he called nominatio, i.e., what the term applies to; and significatio, what hearing the term brings to mind.¹ Jumping ahead closer to Frege’s time, J.S. Mill’s System of Logic, published in 1843, likewise contains a number of ideas familiar to contemporary philosophy of language, such as his distinction between connotation and denotation, his theory of proper names (which holds that proper names are not connotative), and his account of definite descriptions that involves conditions of uniqueness.²

These are just a few of the historical highlights of philosophers’ involvement with semantic themes. A linguistic tradition in philosophy can be traced far back in India and China, too. But, in Europe it was not until the approach of the 20th century that language and meaning came to be regarded as proper subjects of philosophical investigation in their own right. After that, it was not long before problems connected with meaning began to be seen in parts of Europe and the United States as the most fundamental of all philosophical problems. This view was dominant up through the 1960s. It persists among some to this day.

In the history of Western philosophy this focus on language ushered in what is often called ‘the linguistic turn’. Language and logic began to be viewed not only as subject matter comprising a body of knowledge, they were also heralded as a collection of techniques by means of which longstanding intellectual disputes could finally be settled. Enormous claims were made on behalf of this purportedly new approach and, especially, of the impressive toolkit of logical devices and formal techniques that it brought to the job. Many, like Bertrand Russell (1872–1970), seemed to think that all philosophical puzzles were to be solved by linguistic and logical analysis. Others, such as members of the Vienna Circle (1922–ca. 1940), argued that the means of verifying the truth of a declarative form of words not only determined its meaning, but, insofar as it also revealed which forms of words were, despite their superficial character, literally without meaning, it provided a criterion for meaningfulness itself. The early writings of Ludwig Wittgenstein (1889–1951) played a formative role in this; but the figure most responsible for the linguistic turn is Gottlob Frege. The founder of modern logic, Frege is widely hailed as the father of analytic philosophy. It is with the work of Frege, beginning around 1879, that this study is primarily occupied.

It borders on understatement to observe that subsequent philosophers of language have to an overwhelming extent shared an overriding interest in the concepts of reference and sense that Frege introduced. To a somewhat lesser extent – one that mirrors Frege’s own degree of concern with this topic – they have also applied themselves to his notion of utterance-force. Considering the weighty claims as to the significance for semantics (and for science and mathematics) of formal devices developed by Frege and by the likes of Polish mathematician, logician, and philosopher Alfred Tarski (1901–1983), it is not entirely surprising, I suppose, that discussions of tone that do more than scratch the surface are almost non-existent. Slim contributions by Oxford philosophers Michael Dummett and Bede Rundle from between 20 to 30 years ago constitute practically the sole exceptions.³

That said, a new interest in tone, however modest, appears recently to have surfaced. Among those advancing or contesting theses generated by or dealing with one or another example of Frege’s in the past dozen years are Stephen Neale (1999, 2001), Kent Bach (1999, 2006), Tim Williamson (2003), Michael Pelczar (2004), Christopher Potts (2005, 2007, 2012), Eva Picardi (2006, 2007), and Laurence Horn (2008).⁴ I am happy to see that Frege’s Farbung, along with the important issues it raises or impinges on, has finally begun to attract the attention it deserves. These individuals have helped to shape, and reshape, my own thinking on this subject, and some of their ideas appear or are otherwise touched on in the central part of this book. For this I am grateful. It is my hope that one or another thought put forward in the following pages will contribute to the ongoing discussion.

1.2

Frege’s Ingredients of Meaning

Although Frege spent his entire working life at the University of Jena teaching mathematics, he once remarked, “Every good mathematician is at least half a philosopher.”¹ If he had done nothing else besides inventing modern predicate logic with the introduction of quantifiers and variables into his function-and-argument approach to analyzing sentences, Frege’s elevated place in the history of both logic and the philosophy of logic would be assured and his genius stamped and sealed for all time. His contributions, however, are not confined to philosophical logic and mathematical logic; on the contrary, they are of immense importance to the philosophy of language. Frege wrote his first published paper, ‘Funktion und Begriff’, 122 years ago, and his last, ‘Nebengedanke’, 90 years ago.² Despite the passage of time and the many developments that have shaped the course of semantics since, Frege’s ideas loom large today. Dummett not long ago observed, “There is scarcely a live question in contemporary philosophy of language for whose examination Frege’s views do not form at least the best starting point.” So it is with the topic of tone and the issues that it involves.

It is, in my opinion, a great misfortune that Frege’s views on language, as on other matters of logic and mathematics, were not known for some time outside a small circle, a circumstance that was the source of increasing distress and ultimately of considerable bitterness on his part toward the end of his life. It was not until 1949, some two dozen years after his death, that the first English translation of ‘Über Sinn und Bedeutung’ appeared.³ However, once his work became known outside Germany, the influence of his thought spread rapidly. It soon became the central focus of the discipline (of semantics), the dominant influence on many new lines of thinking which it spawned in ensuing decades, whether in favor or in opposition – including that coming from the likes of Ludwig Wittgenstein and Bertrand Russell, originally, followed swiftly by Peter Strawson, John Austin, Paul Grice, and Willard van Orman Quine in the 1950s and 1960s, and prominently by Donald Davidson and Michael Dummett from the 1960s through the 1980s and 1990s, to name but a small handful.

While Frege’s distinction between sense and reference has long since achieved a kind of legendary status, it remains hotly debated; and his function–argument analysis of sentences and his concept–object distinction continue to be carefully worked over. The distinction between sense and force, no less crucial than the one between sense and reference, has also become something of a fixture in semantic theory, as has his celebrated articulation of a thesis of compositionality and the fundamental context principle. These latter two, especially, along with Frege’s anti-psychologistic approach to explaining truth and meaning – an approach motivated in large part by a desire to make these two notions scientifically respectable by rescuing them from the realm of psychology – naturally play an important role in the study of tone.

As a beginning, then, one thing that can usefully be said, although it comes as near as may be to a platitude, is exemplified by the following observation: the sentence ‘Barack Obama is a leading candidate’ differs in meaning from the sentence ‘Barack Obama is a leading suspect’, and differs, specifically, according to the difference in meaning between the words ‘suspect’ and ‘candidate’. Or, to take another equally obvious example, where the addition of a single word makes for a difference in meaning between ‘I have ten dollars in my wallet’ and ‘I have exactly ten dollars in my wallet’. This is just to say that the meanings of sentences are sensitive to even minute differences at the level of individual words and other components. Compare ‘Gates is wealthy and generous’ with ‘Gates is wealthy yet generous’. The overall meaning of a sentence is sensitive not only to different words, but to their arrangement, too: cf. ‘Michael Jordan has officially retired from basketball’ and ‘Has Michael Jordan officially retired from basketball?’

This confirms Frege’s celebrated compositionality thesis concerning sentence meanings, expressible by the equally trite (but nevertheless crucially important) observation that the meaning of a sentence is a function of the meanings of its individual words together with the way in which they are arranged. Although not every difference corresponds to a difference in meaning – cf. ‘Trixie has one blue eye and one green eye’, and ‘Trixie has one green eye and one blue eye’ – the approach I am advocating respects Dummett’s parsimonious attitude toward the assignment of expression-meanings. As a rule one ought to seek explanations that, as far as possible, assign constant meanings to individual expressions across the different linguistic environments and conversational contexts of their actual use. Thus, throughout this study I try to maintain a commitment to the principle of semantic innocence. According to Donald Davidson, a semantic theory for a language is innocent if and only if the semantic values, or meanings, of its expressions do not vary according to the linguistic environment in which they occur.⁴ This idea can and should be broadened to incorporate not only individual words and compound expressions, but linguistic form more generally, including such things as declarative, interrogative, and imperative form. Dwight Bolinger’s slogan, “One form, one meaning”, should be the motto that guides.⁵

On the flip side, it seems often to be assumed that two words or two constructions that differ in form can nevertheless possess the same meaning. In order to facilitate systematization, important distinctions get sacrificed for the sake of theory, and meaning is reduced to the truth-values of formal logic. Another point that favors univocity is Frege’s context principle, namely, the idea that it is only in the context of a sentence that an individual word has a meaning. In other words, a word on its own, so to speak, whether a name, a proper noun, a verb, a preposition, an article, or what have you, says nothing at all. Dummett unpacks this as the thesis that in the order of explanation of our general concept of meaning, the idea of ‘sentence-meaning’ has priority over that of ‘word-meaning’, even though the meanings of individual sentences are determined by the meanings of their constituent words. Another way of putting Frege’s context principle is this: it is only with complete sentences that we can say anything at all. Hence, word-meaning, at its most basic level of conception, is to be explained in terms of the contribution that individual words make to the meanings of sentences in which they are embedded, where this is laid out broadly in terms of the conditions that make for a proposition’s being true.⁶ I accept this version of Frege’s context principle; but it is important to understand it as relating the meanings of individual words not to any particular sentence, or set of sentences, but to all types of constructions in which they occur. One cannot be satisfied that one has framed an adequate account of the English adverb ‘not’, for instance, having looked only at examples for which it can be translated or paraphrased as a truth-functional sentence-operator.⁷

Sometimes syntactic considerations press us to assign different meanings to what appears, morphologically, to be the same word: e.g., ‘bank’ (along the Thames, for walking or fishing) and ‘bank’ (of the National Westminster kind, for conducting financial transactions). With such examples there is a strong inclination to speak, not in terms of two meanings of a single word, but rather, of distinct words. Other cases do not bifurcate: cf. ‘remind’ as it occurs in ‘He reminded me of a former colleague’ and ‘He reminded me of my prior obligation’. Plainly, these two sentences differ in meaning; but we should look to the obvious differences in verbal complementation, rather than to any supposed ambiguity in the verb, to account f...

Table of contents

• Cover
• Title
• Introduction
• Part I   Historical Preliminaries
• Part II   Varieties of Tone
• Part III   Tone and the Representational Character of Meaning
• Notes
• Bibliography
• Index