On March 22, 2002, in Dili, the capital city of East Timor, a national constitution was enacted, whose Article 13 stated that “O tétum e o português são as línguas oficiais da República Democrática de Timor-Leste,” meaning ‘Tetum [a Southeast Asian language] and Portuguese are the official languages of the Democratic Republic of East Timor.’ A significant detail is that all the names and most of the surnames of the parliamentarians who signed the new constitution are Portuguese (Lourdes, Manuel, Maria, José, Luisa, Norberto, Costa, Martins, Silva, Alves, and so on), even though reportedly only about 2% of the population of East Timor speak Portuguese (Ethnologue.com 2002).

Having a constitution was a major accomplishment for that small country. After becoming independent in 1975 from Portugal, whose colony it had been since the fifteenth century, Timor was occupied by Indonesia for the next twenty-five years, and had to secure its freedom again at a heavy toll in human lives. Historically, however, this is just one more occasion on which Portuguese has served not only as a vital link to the outside world but also as a common language for speakers of East Timor's nineteen other languages, some of which, like Adabe or Habu, have only about one thousand speakers each. Portuguese has often played the role of a lingua franca° (a topic to be taken up again in Chapter 6) since the fifteenth century, when it began to spread from its birthplace in the Iberian Peninsula to reach the four courners of the earth.

In Chapter 1 we glanced at the external history of Portuguese, and in Chapters 2 through 4 we examined aspects of its sounds, word forms, and the structure of phrases and sentences. In doing this we followed a synchronic° perspective, looking at the language in its contemporary state, without taking into account historical factors, as though the language were fixed in time.

As time goes by, however, every language undergoes a variety of diachronic° processes that substantially affect all of its components and which constitute its internal history. Such modifications affect sounds (phonetic and phonological change), the shape of words (morphological change), the structure of phrases and sentences (syntactic change), the make-up of the lexicon (lexical change), and the meaning of words (semantic change). Given enough centuries, a language can be changed into a substantially different one. In this chapter we will consider some of the diachronic processes responsible for transforming popular spoken Latin into Portuguese.

The Latin source

Although it is difficult to observe diachronic processes directly, we have sufficient data about Latin and early forms of Romance, provided by ancient manuscripts, inscriptions, and grammarians' comments about the language. By comparing such information with data from today's Romance languages, we can make reasonable conjectures about what happened. A major obstacle, however, is that understanding those processes requires a knowledge of Latin, which most students nowadays lack.

Like any language spread over a large territory occupied by a stratified society, Portuguese as spoken in Brazil encompasses a raft of partially overlapping regional and social varieties that show a significant amount of contrast in pronunciation and syntax. The present chapter will review some of the specific features of Brazilian Portuguese.

Variation in Brazilian Portuguese

A great deal of such variation is directly related to speakers' educational level, which in turn is linked to their socioeconomic situation. In fact, some of the most salient contrasts within Brazilian Portuguese are not regional but social. There is considerable divergence between the vernacular° speech of the majority of the population, the speech of the educated minority, and the normative° language codified in prescriptive° grammars. Traditionally, such grammars have been based primarily, if not exclusively, on the formal written usage found in Portuguese (and, as of the 1920s, also Brazilian) literary works spanning over four centuries. In principle, there is nothing wrong with this, for it is necessary to have reliable descriptions of the language used in literature. The error, however, has consisted in taking the literary variety as being the only valid one, and in condemning varieties diverging from it – such as the vernacular – as the result of decay caused by speakers' poor language habits. In addition, it has been determined that poor theoretical principles and unsystematic selection of examples have led normative grammars to contradict each other (Castilho 1989a:57).

Even if Polonius found Hamlet's reply “Words, words, words” a trifle odd, he had no reason to question what it meant – after all, the prince was holding a book, which is where anyone would expect to find words. Like Polonius, most of the time we act as if we knew what a word is, and feel no urge to question the status of words like chuva ‘rain,’ filhinho ‘sonny,’ ponta ‘point,’ guarda ‘guard,’ pé ‘foot,’ or papai ‘daddy.’ But what about forms like pontapé ‘kick,’ guarda-chuva ‘umbrella,’ filhinho-de-papai ‘mamma's boy’? Are they single words or combinations of two or three words? And what should we say of word combinations that function like a meaning unit, such as bico-de-papagaio, rabo-de-arara, parece-mas-não-é or planta-de-Natal, which despite their literal meanings (respectively ‘parrot's beak,’ ‘macaw's tail,’ ‘it-seems-but-it-isn't,’ ‘Christmas plant’) are simple regional Brazilian names for the Euphorbia pulcherrima, the poinsétia ‘poinsettia’ of Yuletide fame? In this chapter we will look into morphology, itself made up of two Greek words, morphē ‘form’ + logos ‘study,’ that is, the study of the form of words.

Words and morphemes

Despite their variety, words have a definite internal structure. In words like carros or senhores, we recognize two formants. One is the stem° (carr -, senhor), which bears the lexical° meaning of an extralinguistic referent°, such as things, persons, ideas, and so on. The other is a suffix° (-s, -es), attached to the stem and carrying the meaning ‘plural.’

Introduction

Thus Frege (1892c: 152) introduces the sense/reference distinction. A sign, he says, expresses [Ausdrucken] its sense and stands for, refers to, denotes [Bedeuten] or designates [Bezeichnen] its reference. Frege (1892c) confines his discussion to proper names [Eigennamen]; but he intended the sense/reference distinction to apply as well to concept words [Begriffswörter], and to function-expressions generally. A careful reading of Frege's later writings confirms this, but the decisive evidence is to be found in the unpublished manuscript which the editors entitled “Ausführungen über Sinn und Bedeutung”:

Introduction

The matter of existence is one of the most difficult in philosophy. The topic is infused with a particularly noxious mix of dogma and confusion. Needless to say, Frege's influence on modern thought is deep. There are three distinct aspects of the issue on which he made contributions:

• First, there is the Context Principle – “never to ask for the meaning of a word in isolation, but only in the context of a proposition” (Frege 1884b: x) – which he employed to promote his own view that numbers are objects and to undermine the then current psychologism in mathematics;

• Second, there is his treatment of nonreferring singular terms and the truth value of sentences containing them;

• Third, and perhaps most significantly, there is his doctrine that existence is a property of properties, not of things.

We have little to offer that will help clear the general fog about Frege's Context Principle and its application. We spoke about Frege's treatment of nonreferring singular terms in Chapter 3, and again in Chapter 6. We will say more in Section 7.6. However, we will focus in this chapter primarily on the third issue, namely, whether existence is a first-order property.

Frege is widely credited with providing a precise interpretation in the language of modern logic of Kant's (1781: 504) well-known declaration: “‘Being’ is obviously not a real predicate.…” Frege's discussion of the issue is self-consciously derived from Kant's, and framed with the same explicit connection to the Ontological Argument for God's existence.

Introduction

Few texts are as well known to modern philosophers as Frege's (1892c: 151–2) opening paragraph:

Introduction

Frege's story about indirect contexts – ‘that’ clauses, like ‘Harry believes that’ and ‘Joan said that’ – is widely known and enormously influential. And yet it is only the briefest of sketches. Here is what he says in “On Sense and Reference”:

It is quite elegant, the way he has knitted together the sense/reference distinction and the problem of substitutivity in oblique contexts. But it is not as tight as one might think. For we know what the customary reference of an expression is supposed to be: the customary reference is the thing the word stands for. We also know what the customary sense of an expression is supposed to be: the customary sense is, roughly, the meaning of the expression. So we know what the indirect reference is supposed to be: Frege explicitly identifies the indirect reference with the customary sense. But what is the indirect sense supposed to be?

Introduction

We saw in Chapter 4 that Frege had applied the Begriffsschrift surrogate for identity, identity of content, to sentences as well as to names. Frege (1879) believed that names and sentences both stood for their contents. Frege (1892c) saw no need to change his treatment of sentences as names. The issue he addressed was not whether sentences refer, but what they refer to. He sought to correct his early account – as well as related views which take propositions, thoughts, states of affairs, or facts as the items designated by sentences. These items, he now thought, belonged at the level of sense. In a very influential argument, Frege defended the view that the two truth values – true and false, or as Frege preferred, the True and the False – are the only candidates that are functionally related via the Compositionality Principle 2.3.1 to the reference of the parts of the sentence, and which, in turn, are functionally related to the reference of larger constructions in which the sentences are embedded. “If we are dealing with sentences for which the Bedeutung of their component parts is relevant,” Frege (1892c: 158–9) asked, “then what feature except the truth value can be found that belongs to such sentences quite generally and remains unchanged by substitutions of the kind just mentioned?” In this chapter, we will examine carefully his views on truth, and, in particular, raise doubts about the inevitability of this result.

Introduction

We saw in Chapter 3 that Frege and Russell chose different strategies to deal with the Paradox of Identity. The problem for each was the informative character of definite descriptions. Frege (1892c) continued to regard both ordinary proper names as well as definite descriptions as belonging to the same syntactic category: both were Eigenname. He identified the informativeness of these expressions with the sense they expressed, but he does not appear ever to have attempted to link up this sense he attached to an Eigenname in any systematic way with the semantic role of predicate expressions. Is the sense attached to a proper name to be identified with a concept, or a combination of concepts, denoted by some corresponding predicate? This does not seem right, for concepts are extensionally equivalent while senses are not. Is the sense attached to a proper name to be identified with the sense of a predicate – and if so, how? These are issues Frege simply did not address. Russell (1905), however, met these issues head-on. He took the informativeness of definite descriptions as evidence that they were predicative in nature: he regarded their status as singular terms as a surface feature of what are at bottom, logically predicative constructions. Russell had a much more comprehensive theory of definite descriptions than Frege did, one that so struck the philosophical community that its very methodology served, in Ramsey's words, as a “paradigm of philosophy.”

Introduction

Many commentators have been content to accept Frege's (1892c) account of his Begriffsschrift theory of identity, resulting in a somewhat distorted picture of the sense/reference theory. What little criticism there has been of the Begriffsschrift view can be grouped into the following three charges:

1. It has been alleged that the information contained in an identity statement, when interpreted in the manner of Begriffsschrift, can only be the trivial information that the linguistic community has adopted such-and-such conventions, not the substantial information embodied in a genuine discovery about the world. (This is derived from Frege's (1892c) own criticism of the Begriffsschrift theory.) See Linsky (1967), Kneale and Kneale (1962).

2. It has been alleged that the Begriffsschrift theory is circular or that it involves a vicious infinite regress. See Russell (1903b), Wiggins (1965), Kneale and Kneale (1962).

3. It has been alleged that the Begriffsschrift theory is flawed by use/mention confusion. See Church (1951), Furth's introduction to Frege (1893).

Not one of these adequately reflects the subtlety of Frege's Begriffsschrift view. We will examine and evaluate what he says in Begriffsschrift in Section 4.2, and then turn to these three criticisms: we treat the first in Sections 4.3 and 4.4, the second in Section 4.5, and the third in Section 4.6.