Read an ExcerptA Brief History of the Spanish Language
By DAVID A. PHARIES
The University of Chicago Press Copyright © 2007 David A. Pharies
All right reserved.
Chapter One Language Change
The Inexorability of Language Change
As mentioned in the introduction, given the nature of reality as we understand it, it is not at all surprising that human languages should be subject to an irresistible process of change. What would be surprising is if they were not.
To say that language change is inexorable and irresistible suggests that it is also unstoppable, although there have been numerous attempts throughout the ages to stop it or at least to slow it down. This explains, for example, the existence of a well-known genre of corrective reference works, with titles like Diccionario de dudas ('dictionary of doubts') or Diccionario de incorrecciones ('dictionary of mistakes'), that purport to help users defend themselves against the waves of change that incessantly assault the language. In one such work we are warned, for example, against saying or writing la ciudad adolece (should be: padece) de escasez de agua 'the city suffers from a water shortage', enfrentar (should be: afrontar) un reto 'to face a challenge', el agudizamiento (should be: la agudización) del problema 'the increasing severity of the problem', al cabo (should be: al final) de la reunión 'at the end of the meeting', no tenía otra alternativa (should be: opción) 's/he had no choice'. The resistance to language change by the French is particularly strong, as they have even been known to threaten users of Anglicisms, i.e., loanwords or borrowings from English, with monetary fines.
Changes in Progress
Thanks to sociolinguistics, it is now recognized that the processes whereby changes are introduced and eventually generalized in natural languages is susceptible of direct observation. Seen from this perspective, such changes are termed changes in progress.
Certain kinds of changes are not particularly difficult to perceive, such as lexical or vocabulary changes, which include the introduction of new words. Spanish speakers everywhere will have noticed the appearance of such neologisms as fax 'fax', píxel 'pixel', escáner 'scanner', genoma 'genome', lifting 'face-lift', and puénting 'bungee-jumping'. Semantic changes-changes in meaning-are more difficult to detect, such as the use of retirar 'to withdraw' for jubilar 'to retire', or of atender 'to attend to' for asistir 'to attend', both probably due to the influence of their English equivalents. In the case of phonetic changes, i.e., changes in pronunciation, speakers tend to detect them unconsciously: They know that there is something different about the way someone speaks, but they cannot identify it precisely. They say, "That's the way young people talk these days," or "He sounds like a surfer".
With a little practice, it is possible to detect incipient morphosyntactic or grammatical changes. In Spanish, many speakers have noticed the tendency to make the verb haber agree with its subject in certain impersonal expressions, as in habían (should be: había) dos coches ante la casa 'there were two cars in front of the house', and habrán (should be: habrá) varios tipos de comida en la fiesta 'there will be several types of food at the party'. Conversely, agreement is eliminated in the use of the singular clitic pronoun le for plural les: Le (should be: les) escribo a mis amigos 'I write to my friends'. The above-mentioned tendency to use que for de quien in sentences such as es el tío que (should be: de quien) te hablé ayer is another change in progress. In American English the longstanding tendency to confuse subject and object case in pronominal use continues unabated. Probably only English teachers, for example, can still distinguish with confidence between subject who and object whom. The confusion is especially acute where two pronouns, or a noun and pronoun, occur together. The use of object pronouns in the subject slot has been around for a long time, cf. sentences such as me and him (should be: he and I) went to the movies or me and my brother (should be: my brother and I) went to the movies. In contrast, the inverse usage, that is, of subject pronouns in the object slot, is recent and quickly becoming generalized, cf. they sent it to my brother and I (should be: my brother and me) and they saw she and I (should be: her and me) as we left the apartment. Another change in progress is the tendency to replace past participles with past tense forms: They have already went to school, I have ran out of gas twice this year.
Language Change as Observed through Written Evidence
Before the advent, at the end of the nineteenth century, of the technology for electronically recording sounds, the only way to learn about previous language states was through written documents. It is thanks to writing that we have the opportunity to study older stages of a languages's development, but, at the same time, we must recognize that the written medium imposes certain restrictions. Any document that has merited preservation for a long time, even centuries, will inevitably reflect the most formal style or register of the language. This means that we will forever be denied access to the everyday colloquial speech of the distant past. It is necessary also to keep in mind that orthography does not always reflect pronunciation faithfully. Some scholars believe, for example, that around the beginning of the second millennium it may have been customary to read Medieval Latin as though it were the vernacular, pronouncing homine 'man' as [ómbre] and amicos 'friends' (or even amici, the Latin nominative plural form of amicus) as [amígos]. The hypothesis presupposes a radical separation between writing and speaking that would seem absurd if it were not confirmed by the cases of Modern English and French: Pronouncing homine as [ómbre] is not qualitatively different from pronouncing knight as [náit] or temps as [tã].
In spite of the difficulties that the interpretation of written texts presents for historical linguistics, it is fascinating to see the cumulative effect of centuries, even millennia, of linguistic change, reflected in ancient texts. Comparisons between various stages of a language are particularly revealing when the written documents involved represent chronologically distinct versions of the same content, as in the example that follows: three versions of the biblical creation story.
Vulgate (Fourth Century) In principio creavit Deus caelum et terram. Terra autem erat inanis et vacua, et tenebrae erant super faciem abyssi, et spiritus Dei ferebatur super aquas. Dixitque Deus: Fiat lux. Et facta est lux. Et vidit Deus lucem quod esset bona et divisit lucema tenebris. Appellavitque lucem diem et tenebras noctem. Biblia Romanceada (Thirteenth Century) En el comienço crio Dios los cielos e la tierra. E la tierra era vana e vazia, e la escuridat sobre la faz del abismo, e espiritu de Dios ventiscaua sobre fazes de las aguas. E dixo Dios: sea luz, e luz fue. E vido Dios la luz que era buena; e aparto Dios entre la luz e entre la tyniebla. E llamo Dios a la luz, dia, e a la escuridat llamo noche. Modern Spanish Bible Al principio creó Dios los cielos y la tierra. La tierra estaba confusa y vacía y las tinieblas cubrían la haz del abismo, pero el espíritu de Dios se cernía sobre la superficie de las aguas. Dijo Dios: Haya luz, y hubo luz. Y vio Dios ser buena la luz, y la separó de las tinieblas. Y a la luz llamó día, y a las tinieblas noche.
Our first impression is that some very fundamental changes must have occurred since the Latin stage, as the first text is largely undecipherable for the speaker of Modern Spanish. Still, a comparison of the texts allows us to reach certain conclusions about the earliest form of the language. For example, since there appear to be no Latin words corresponding to the definite articles in the modern text, it can be assumed that Latin has none. We also note that nouns seem to have more than one form, cf. tenebrae /tenebris /tenebras, also Deus /Dei, lux /lucem, and terra /terram. Regarding verbs, we suspect that creavit might be associable with Sp. crear 'to create' and esset with Sp. ser 'to be'. On the other hand, certain forms remain incomprehensible, such as ferebatur and fiat.
The thirteenth-century Spanish, in contrast, is easy to decipher. The ç of comienço (mod. Sp. comienzo 'beginning') and the x of dixo (dijo 'said') are somewhat odd but do not impede comprehension. Neither do the orthographic differences in vazia (for vacía 'empty'), tyniebla (for tinieblas 'darkness'), and escuridat (for oscuridad 'darkness') prove particularly vexing for the native speaker of Modern Spanish. Vido, as well, is easily associable with vio 'he/she saw', and ventiscaua is clearly related to viento 'wind'.
This little comparative exercise gives us an idea of the phenomena that we will be dealing with in the chapters to come. It also serves to illustrate, quite clearly, the fact that when we speak of Latin, Medieval Castilian, and Modern Spanish, we are speaking of a single linguistic entity whose existence extends throughout this period of time. Thus, from the historical point of view, modern-day speakers of Spanish would be justified in calling their language Latin. If they do not do so, it is because Spanish is not the only living descendant of Latin: French, Italian, Sardinian, and several other languages can also make this claim. Concomitantly, it is clear that when we say that Latin is a dead language, we are speaking only of a certain temporal variety of a language that, in other guises, continues to be spoken.
The Categories of Language Change
Linguistic changes can be classified according to the linguistic component they affect. Thus, we recognize changes as being phonetic, phonemic, morphological, syntactic, semantic, or lexical, as illustrated below.
Phonetic: [t[??]] is depalatalized in some dialects or varieties of Modern Spanish, approximating [ts], so that chico 'small' is pronounced [tsíko]
Phonemic: /[gamma]/ merges with /j/ in some dialects of Modern Spanish, making homophones of minimal pairs (sets of two words with different meanings that differ in only one phoneme) such as halla 's/he finds' and haya 's/he may have'
Morphological: Lat. cedunt produces ceden 'they yielded' in preliterary Spanish instead of the expected cedon, through the influence of verb forms such as deben 'they should', which derives regularly from Lat. debent
Syntactic: miráronse changes to se miraron 'they looked at themselves/ each other'
Semantic: falda 'skirt' acquires the additional meaning 'mountain slope' in Modern Spanish
Lexical: the archaic conjunction ca 'because' is lost in the sixteenth century, in favor of its synonym porque
In general, the changes in each category are unique in character and obey different principles. However, there can be important links between categories, since sometimes a change in one category will bring about or necessitate a change in another category. For example, Latin grammar employs morphological endings to indicate the functions of nouns within sentences-e.g., to indicate whether a noun is acting as subject or object of a verb. It is possible that the loss of this system of endings (which may have been favored by certain phonetic processes of weakening and loss) is the motivation for a syntactic adjustment whereby the functions of nouns are signaled in other ways, such as the more frequent use of prepositions or a less flexible word order.
The Causes of Language Change
Several theories have been proposed to explain why languages change.
During the first half of the twentieth century, it was customary to search for an external explanation for every change, based usually on the supposed interference of a language in contact. It was hypothesized that changes were introduced by bilingual speakers who confused their two languages and transferred linguistic features from one to the other. Now, this explanation appears plausible in situations where one language is practically being replaced by another-i.e., in instances of language shift. For example, Reyes 1982 cites a series of hybrid structures in Chicano Spanish that are obviously due to the interference of English, the dominant language in the United States. Examples include syntactic changes such as hizo improve mucho (for mejoró mucho 's/he improved a lot'), which exemplifies the syntactic construction "hacer 'to do' + English infinitive", and los están busing pa otra escuela (for los transportan a otra escuela en autobús 'they are busing them to another school'), which illustrates the construction "estar 'to be' + English gerund". In word formation, we see cases such as taipear 'to type', a verbal derivative whose stem is Eng. to type, also puchar 'to push' (< to push) and mistir 'to miss' (< to miss), as in mistir un tren 'to miss a train'.
However well the interference hypothesis explains cases such as these, it is not valid in situations where bilingualism, i.e., the habitual use of two languages, is less intense or exists only at the margins of society, since it is improbable that monolingual speakers of a language would change their way of speaking to imitate such bilinguals, whose strange form of speaking would hardly be considered prestigious or worthy of imitation. This consideration invalidates the great majority of the hypotheses of this type, such as those proposed by Fredrick Jungemann (1955) to explain the phonological eccentricities of Spanish vis-à-vis other Ibero-Romance languages.
Another explanation for language change is based on the principle of least effort, according to which the tendency of speakers of a language to exert as little effort as possible in expressing themselves entails an inexorable process of degradation. Clearly, however, the tendency to minimize effort is tempered by the need to convey the content of one's utterances with sufficient clarity to avoid misunderstandings or the need to repeat. In any case, the final effect of the uncontrolled degradation of a language would be the loss of its communicative power.
Some scholars have proposed that the cause of language change is imperfect language learning in children. It is true that children must develop their linguistic competence based on data that are at once incomplete (since no child hears all possible utterances) and often imperfect (since a significant proportion of utterances contain errors-the result of lapses, ignorance, speaker fatigue, or other causes). However, even though this explanation might be valid for the idiolect of a single child, it is difficult to see how it could explain changes in a whole speech community, since each child hears different utterances and different mistakes. It would be logical, under the circumstances, to expect the eccentricities in individual speech to cancel each other out in the broader linguistic community.
A closely related hypothesis identifies the inherent heterogeneity of language as the cause of language change. It is well known that utterances are inherently variable, in the sense that no one utterance is precisely identical to another, due to more or less minute differences in the articulation of the sounds, the melody of intonation, the selection of morphemes, etc. According to this theory, the random accumulation of a single variant might result in the generalization of a change. For example, if a barely palatalized variant of the consonant [t[??]] should arbitrarily become dominant, a new articulation [ts] could establish itself as the new norm. The problem with this explanation is that it does not identify the factor that would initiate or condition the selection of the less palatalized variant. Under normal conditions, given the variability of language, one might expect barely palatalized variants to be canceled out by very palatalized variants.
Excerpted from A Brief History of the Spanish Language by DAVID A. PHARIES Copyright © 2007 by David A. Pharies. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.