Many working syntacticians or phonologists do not care about what all linguists tend to call philosophical issues, such as “What kind of object is a language?” They are happy to leave the ontological basis of their work unanalyzed. It is possible to accept this situation. We may choose to inquire directly and separately in what way pieces of external evidence from Acquisition, Borrowing, Change, and Damage may have an impact on ideas considered seriously by working linguists. However, the moment we decide to group the ABCD of these four domains together under the E of Externality, thus trying to organize them into an area of inquiry, this decision leads naturally to a more general type of question. Unless we grapple with such a question, our ABCD list begins to look like the well-known Borges quotation from a fictitious Chinese encyclopaedia stating that animals are divided into:

Progress in linguistic theory during the past 20 years has made it increasingly clear that language acquisition must be viewed as an essentially deductive process in which the child analyzes the input data s/he is exposed to on the basis of an innately specified set of restrictive principles — technically known as Universal Grammar — which narrowly constrain the kinds of hypotheses a child will consider vis-à-vis a given set of data (cf. Chomsky 1980, 1981, 1986; Hornstein and Lightfoot 1981; White 1982; Felix 1987). As a consequence, there is a growing interest in the question of how exactly principles of Universal Grammar interact with the child’s linguistic experience during the course of language acquisition (see e.g., Pinker 1984; Hyams 1986; Lust 1986b; Roeper and Williams 1987 among others for some more recent proposals). It appears that there are currently at least two competing views about the nature of this interaction. One of these views which I shall call “perceptionism” holds that the task of Universal Grammar (UG) is essentially restricted to constraining the types of intermediate grammars that the child will construct, while the developmental process itself is essentially data-driven, i.e., driven by the child’s (changing) perception of the external evidence. The other view which may be termed “maturationism” claims that UG is both responsible for the types of (intermediate) grammars that in principle may emerge and at the same time for the specific nature of the developmental process. Under the maturationist view language acquisition is therefore seen as a process that is driven primarily by internal, i.e., biologically determined maturational mechanisms.

Years ago, Piaget (1926:121) reported the fact that children, in their spoken language, show little concern for identifying the appropriate antecedents of their pronouns. In the texts he presented, they would typically introduce one or more antecedents followed by a series of pronouns. Such examples show a simple rule of linear order whereby a pronoun needs to refer to an antecedent somewhere preceding it. Further, this rule appeared to characterize the productive language of older children in the years five to ten, when much of the structure of language has already been acquired.

In this paper I address the question what Media Lengua can tell us about the distinction between stems and affixes, the nature of lexical entries, and the relation between the lexicon, syntax, and phonology. It is part of a much larger attempt, coming from a variety of subdisciplines to provide what is sometimes termed external evidence for particular assumptions about the nature of the human language faculty.

As far as phonology and morphology are concerned, the available evidence indicates that the role of L1 in shaping interlanguage is confined to those of its rules that are needed to account for its global alternations, alternations that are independent of its morphology (cf. Cearly 1974, Dressler 1985, Kilbury 1981, Singh and Ford 1982, 1987, Singh and Martohardjono 1989, Wode 1978, and Wurzel 1977, among others). The rules needed to account for the local, morphologically dependent alternations of L1 or the ones needed to account for its word-formation processes do not play such a role. Interference, in other words, can be caused only by across-the-board phonological rules of L1. So-called morphophonemic rules of L1 do not cause it, and morphological interference from L1 seems not to exist as word-formation errors in intermorphology are the results of illegal extensions of L2 word-formation rules (cf. Singh 1989 and Singh and Martohardjono 1989). The purpose of this note is to critically examine the accounts contemporary theories of phonology provide of this state of affairs and to argue that the account provided by the sort of theory proposed in Ford and Singh (1983, 1985a, 1985b) and Singh and Ford (1982, 1987) is the most satisfactory one.

Does phonological conditioning have a privileged status in allomorphic alternations? Most synchronic descriptions appear to assume that if simple phonological conditions can account for some alternation, they automatically constitute the relevant factors chosen by speakers in their internalized grammars (but cf. Hooper 1976:87 or Klausenburger 1979:29–36 for divergent views). Intuitive notions of simplicity cannot provide a solution because simplicity is actually an empirical issue (cf. Chomsky 1965:37–47). The answer can only come from the analysis of external evidence, i.e., errors, language acquisition, historical changes, etc.

L’application d’une théorie linguistique, essentiellement élaborée à partir de données internes, à des données externes (diachronie, acquisition, performance, pathologie, traitement informatique, etc.) et sa vérification à partir de ces dernières pose de nombreux problèmes. Le premier de ces problèmes est celui de l’exportation des prédicats de la théorie dans un domaine autre (et généralement plus englobant) que celui pour lequel ils ont été définis.

Linguistic analyses of aphasic patients’ performances have developed to the point where interesting, non-trivial descriptions and explanations of pathological performances are available in linguistic and processing terms. The purpose of this paper is to present a short synopsis for linguists of several recent analyses of aphasic patients’ disturbances in the sphere of comprehension of syntactic form. This presentation is designed to illustrate some of the logical and methodological issues regarding the analysis of aphasic data, as well as to present specific results of our investigations. On the basis of our case studies, we also suggest several aspects of syntactic structure and parsing.

To an outsider looking at linguistic theory, the problem of internal vs. external evidence seems to arise out of a conflation of two relatively distinct enterprises. The first of these tries to establish a claim that there is a distinct, organic, language faculty (a language organ), that is, a species-specific characteristic common to all humans. The second is an attempt to abstract from the facts of diversity a structure that has equivalent universal characteristics and thus can be seen as an ideal type or as an ideal initial state condition (of a pure and uniform experience) under which a language is acquired by a native speaker. As I understand it, the relationship between these two enterprises is yet to be fully determined given the fact that the physical mechanism that has been deduced to correspond to a property of mind is as of now largely unknown.