Journal of Historical Pragmatics - Volume 13, Issue 1, 2012

Present-day Dutch has two entrenched “grammatical” hearsay evidentials: a construction with zou (originally the past tense form of the verb zullen, cognate with German sollen) and a construction with schijnen (literally, ‘seem’). The closest English equivalent of both constructions is the “evidential nominative and infinitive” (NCI), which pairs an evidential meaning with the morphosyntactic pattern [SBJ be Xed to Inf]. This is a highly productive construction in English, the most typical instantiation of which is be said to. Present-day Dutch has an NCI construction as well, but the lexical possibilities of this construction are limited to a handful of cognition verbs, which — in their NCI use — encode deontic rather than evidential meanings. On the basis of historical corpus data, this paper shows that the Dutch equivalent of English be said to, i.e. gezegd worden te, looked ready at one time to become entrenched as a substantive hearsay construction as well. This paper traces its evolution and explores the questions of why the pattern disappeared and why Dutch, unlike English, did not develop a schematic evidential NCI construction.

This paper uses linguistic theories of politeness to shed light on the functioning of certain discourse features of the Early Middle English Ancrene Wisse. It has been suggested that Ancrene Wisse is generically impolite because of the presence of conventional misogynistic tropes; however, the nature of the text as a religious rule requires a stance in which face-threatening acts, such as admonishing, advising, and even condemning, are common. The text also exhibits clear politeness strategies, including affectionate direct addresses of readers and comments that explicitly exclude readers from the text’s admonishments. The pragmatic effect of integrating these politeness strategies with the genre’s regulatory requirements creates a complex discourse field for readers to compass but, in fact, may provide an ideal environment for the text and its readers to pursue the genre’s Christian ideals.

It is well known that demonstratives are the cross-linguistically common source of third person pronouns due to the functional similarity between them. For this reason, they are morphologically related to or formally indistinguishable from one another in many languages. First and second person pronouns, on the other hand, typically have historical sources other than demonstratives. However, unlike the close relationship between demonstratives and third person pronouns, the fact that demonstratives and first/second person pronouns have a very tenuous diachronic relationship has not attracted much attention in previous studies. Based primarily on historical data from Japanese, the present study shows that there are at least three functional reasons why demonstratives do not usually give rise to first/second person pronouns. This study also discusses a limited context in which a demonstrative does develop into a second person pronoun.

This paper explores the semantic change of the polysemous Japanese word ato from the perspective of metonymy. In order to define how metonymy contributed to ato’s semantic development, I employ Peirce’s conception of the sign. I argue that the rise of ato’s meanings reflects the relations between the components of Peirce’s semiotic model; that is, representamen, object and interpretant. This account challenges a major previous study by Heine, Claudi and Hünnemeyer that approaches semantic change from the perspective of the conceptual notion “metaphor from metonymy”, putting two aspects, namely family resemblance and unidirectionality, to the fore. While I concur with Panther, who identifies metonymy as an indexical (pointing-to) operation, it will be shown that the crucial factor in ato’s semantic change is that indexicality operates in a triadic fashion facilitated by various cognitive processes. That is, Peirce’s “thirdness”, the concept that a sign mediates between the interpretant and its object, is the key to an account of ato’s semantic change.

In Ancient Greek, an impolite order can be uttered by means of a negative interrogative in the future tense (οὐκ ἐρεῖς; ‘Won’t you talk?’). The aim of this paper is to understand to what extent this type of utterance is impolite, and to explain how such a conventional and indirect order can frequently take on an impolite meaning. For this purpose, data are taken from classical drama (Aristophanes’ and Euripides’ plays).Drawing on criteria put forward by recent work on impoliteness, this study provides an accurate description of uses in discourse, in order to establish that this conventional order is never used with a polite intention, but regularly as an impolite order. Impoliteness can be explained by the locutionary form which gives an orientation to the interpretation of the utterance: an indirect and conventional expression cannot be polite if the locutionary meaning is opposed to it.

This paper provides an in-depth discussion and evaluation of different accounts of the historical development of possessive HAVE GOT. Accounts which postulate the insertion of got as a pattern preserver after phonologically reduced instances of possessive HAVE (Jespersen 1931; Crowell 1959; Krug 2000) are problematic because there is no historical evidence for phonologically reduced possessive HAVE. Another line of research argues that possessive HAVE GOT started out as an inference ‘continuation of possession’ in contexts of present perfect HAVE got(ten) ‘have received/have acquired’ (Johnson 1773; Visser 1973; Gronemeyer 1999).A principled account of this development is provided in terms of the conventionalisation of conversational implicatures. The inference ‘continuation of possession’ is a conversational implicature which meets Levinson’s criteria of cancellability, nondetachability, reinforceability and calculability. Conventionalisation of this conversational implicature is evidenced by uses of HAVE GOT in contexts of inalienable possession where the meaning ‘have received/have acquired’ is no longer possible.