Skip to main content
Log in

[Letters to Editor]

  • Letter
  • Published:

From Nature

View current issue Submit your manuscript

Abstract

IN a collection of sound film registrations I find quite a number of regions that resemble the one reproduced in Fig. 1. Such a region consists of a series of bits of vibration with many peaks of different heights and periods of repetition and several vibratory movements of less sharp character, all of them beginning strong and fading to zero. Each bit of vibration can be characterised by its duration and by the systems of peaks and less sharp movements with their heights, periodicities and decrements. An entire region will be characterised not only by all the measurements for the individual bits but also by the numbers for their differences and order of sequence. Every region will have a more or less similar, but yet different, set of numbers for each of its characteristics. For the group formed from the similar cases there will be for each characteristic an average value a, an average variation v and a limit of variation l. For the whole set of i characteristics there will be the averages a1, a2,, ai, the average variations v1, v2,, vi, and the limits of variation l1 l2,, li. The term phoneme might be used to indicate: (1) the whole group of speech sounds with characteristics specified in this way, or (2) a speech sound defined with averages of these characteristics, or (3) a speech sound that may be any one of the group with these characteristics. The choice depends on convenience and accuracy. If we say that a person pronounced the word indicated in print by hatch with the phoneme æ, we do not mean that he used a whole group or an average; we can only mean that he used one of a group. If, however, we say that the speech of a person or a dialect is characterised by the use of the phoneme æ, any one of the three meanings is appropriate.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

SCRIPTURE, E. [Letters to Editor]. Nature 136, 644–645 (1935). https://doi.org/10.1038/136644c0

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1038/136644c0

  • Springer Nature Limited

Navigation