Model(s) of spoken language in 18th-century English pronouncing dictionaries? - Université Paris Cité Access content directly
Conference Papers Year : 2014

Model(s) of spoken language in 18th-century English pronouncing dictionaries?


When it comes to the the representation of English, existing alphabetical systems have periodically been deemed inadequate and/or inaccurate, and this awareness of the discrepancy between written and oral forms have led “linguists” (in the broadest sense) to devise alternatives. Starting with Orrm, in the twelfth century, spelling reformers have sought to make graphic forms match up with spoken ones, in an idiosyncratic and somewhat haphazard fashion (see Scragg 1974). Their spontaneous metadicourse is lacking in any sophisticated criticism: theirs is simply a negative intuition that the current system fails as a device to record spoken language. However, spelling reformers, right up to the sixteenth and seventeenth centuries, never questioned the basic premise of alphabetical notation, but sought only to improve it. In the early eighteenth century, the appearance of pronouncing dictionaries signaled a key rethinking of the old model, with the invention of “respelling”: a representation of language that was graphic, but distinct from spelling; free from that constraint, it became a system flexible enough to show pronunciation, and even to record disagreements among orthoepists regarding pronunciation. From the start, these early transcriptions included information beyond the segmental: for the first time, modifying the traditional orthography was considered not only undesirable, but insufficient to render the complex patterns of speech; eighteenth-century orthoepists initiated a key rethinking of the segmental model of spoken language prevalent since Antiquity. A significant step consisted of formally expressing the principle of bijectivity; spelling reformers had intuited the problem, but they did not explicitly express the idea of “one sound = one spelling” (Beal 1999). In his 1761 Dissertation, Thomas Sheridan did exactly that; and though his own system for respelling did not achieve this ideal, he inspired Thomas Spence, whose Grand repository of the English language is probably the first to use a properly bijective transcription key. The first innovation of pronouncing dictionaries, much earlier in the century, was actually the marking of stress, in Thomas Dyche's Dictionary of Words (1723). This breakthrough is obviously linked to the specificity of English as an stress-timed language; it is revolutionary in the realization that suprasegmental information plays a crucial role in showing pronunciation. The other pivotal innovation consisted of indicating syllable boundaries (in a limited capacity in Bailey 1740; more systematically in Kenrick 1773): the incorporation of syllabification in the respellings is also linked to a particular feature of English, i.e. the phonotactic constraints on vowels that depend on syllable structure. The inclusion of this suprasegmental layer of information in transcription led certain authors to rethink the nature of the segments themselves, by proposing the possibility of ambisyllabicity for intervocalic consonants. Two of the most influential orthoepists, Sheridan and John Walker, were more famous in their lifetimes for their work on elocution, or what we would now call prosody. What, if any, is the link between this suprasegmental level and the segmental, syllabic, lexical analysis of their respellings?
VPouillon_2014_ModelsSpeech.pdf (1.16 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01378592 , version 1 (10-10-2016)


  • HAL Id : hal-01378592 , version 1


Véronique Pouillon. Model(s) of spoken language in 18th-century English pronouncing dictionaries?. Colloque SHESL-HTL 2014: "Modèles et modélisations en sciences du langage, de l'homme et de la société", SHESL (Société d'Histoire et d'Epistémologie des Sciences du Langage); HTL (UMR 7597, Histoire des Théories Linguistiques); Centre Alexandre Koyré (UMR 8560), Jan 2014, Paris, France. ⟨hal-01378592⟩
78 View
42 Download


Gmail Facebook X LinkedIn More