Syntagmatic Structures. Distribution and Probability

  • Bertil Malmberg
Part of the Kommunikation und Kybernetik in Einzeldarstellungen book series (COMMUNICATION, volume 2)


Any one of the minimal units set up by means of the segmentation and the phonemic analysis described in the preceding chapters gets its linguistic value through its relations to the other units of the paradigm. Its amount of information is to a large extent determined thereby. The richer the system, the greater the amount of information of any of its units, and vice versa. But as a member of a given chain of expression units, i.e. in a syntagm, it necessarily also has given relations to any other units within the chain. These are the syntagmatic relations to which this chapter will be devoted. We shall soon see, by means of a few examples, that even the structure of the syntagm, and, more generally, of the complex expression units (syllables, groups, etc.) are of importance for the informative value of the phonemes. The laws of distribution valid for the building up of these larger units often strongly reduce the amount of information of each smaller unit. In no language is the combination of the phonemes, or more generally the putting together of smaller entities into larger elements, free. In most languages, it is subject to very strong limitations.


Word Boundary Expression Unit Syllabic Structure Vowel Harmony Phoneme Sequence 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliographical Notes

  1. The idea of consonants as marginal units and of vowels as central units is old, reflected e.g. already in the term “consonant” (“which sounds with something”), or in the German terms “Selbstlaut” and “Mitlaut” (Swedish “självljud”, “medljud”). Traditional handbooks often define consonants as sounds which cannot alone make up a syllable. — The idea that the syllable is just a phonological (structural) unit is due to Otto von Essen (works quoted in Chap. IV). See also his chapter on the syllable in the 3rd edition of his “Allgemeine Phonetik” (quoted above). Cf. my criticism in the articles quoted in Chap. VI. Carl Hj. Borgstrøm published his article in “Norsk Tidsskrift for Sprogvidenskap” XV, 1949, pp. 137 – 187. I have published an article “Voyelle, consonne, syllabe, mot” in “Miscelanea Homenaje a André Martinet”, part III, 1962. The generality of open syllables has been pointed out by Jakobson (works quoted above). For the primitive word type mama and papa, see Jakobson, “Why ‘mama’ and ‘papa’ ?” in “Perspectives in Psychology”, 1961, pp. 124–134. For Japanese syllabic structure, see e. g. G. Wenck, “Japanische Phonetik” I, 1954, §§ 54–55. For Sigurd’s studies, see “Studia linguistica” XI, 1955, and XII, 1958; cf. his “Ponotactic Structures” 1965;Google Scholar
  2. Eli Fischer-Jørgensen’s in “Acta linguistica” VII, 1952, pp. 8–39. The mathematical theory behind Sigurd’s analysis was exposed by Lars Gårding in “Studia linguistica” XI, pp.21 – 34 (“Relations and Order”). The syllabic structure of Spanish was treated by the author in “La structure syllabique de l’espagnol” /Boletim de filologia IX, 1949, pp.99–120/, and in “Notes sur les groupes de consonnes en espagnol” /Zeitschrift für Phonetik II, 1948, pp.239–255/; cf. also my “Linguistique ibérique et ibéro-romane” /Studia linguistica XV, 1961, pp. 57 – 113/. The substitution of vowel length or vowel quality for final -s in Spanish dialects was treated by T. Navarro Tomás., “Travaux du Cercle linguistique de Prague” VIII, 1939, pp. 184–186; and E. Alarcos Llorach, “Fonología y Fonética” /Archivum VIII, pp. 191–203, Universidad de Oviedo, Facultad de Filosofía y letras, 1958/. Hottentot phonetics is described by Beach in “Phonetics of the Hottentot Language”, 1938.Google Scholar
  3. The example given is quoted from G. S. Nienaber in “Album Edgard Blanquaert”, 1958, pp.113 – 122.Google Scholar
  4. For the phonetics of Modern Greek, see J. T. Prings, “A Grammar of Modern Greek on a Phonetic Basis”, 1952. — The phonetic unity of the word was treated by Björn Collinder in “Das Wort als phonetische Einheit” /Sprâkvetenskapliga sällskapets i Uppsala förhandlingar 1937 – 39, pp. 63 ff./. The “word” was treated by A. Rosetti in “Linguistica”, 1965. The French preference for open syllables has been proved by Gunnar von Proschwitz, “Étude sur la répartition des syllabes ouvertes et fermées en français moderne”, 1953.Google Scholar
  5. For the background of consonant alternation in Celtic, see e.g. Holger Pedersen, “Vergleichende Grammatik der keltischen Sprachen” I, 1909, Chap.XIII, §295ff.; and Martinet, “Économie”, pp. 257ff. For Franz Giet’s studies on Chinese tones, see his thesis “Zur Tonität nordchinesischer Mundarten”, 1949. Menzerath’s contributions to the structure of German words were given in “Studia linguistica” IV, 1950, pp. 54 – 93 (together with W. Meyer-Eppler) and in his book “Die Architektonik des deutschen Wortschatzes” /Phonetische Studien, herausgeg. von Paul Menzerath. III, 1954/. Adolf Noreen’s phoneme statistics for Swedish were published in “Vårt språk” I, 1903, p. 542. G. K. Zipf’s fundamental works were “Relative Frequency as a Determinant of Phonetic Change” /Harvard Studies in Classical Philology XL, 1929/, and “Human Behavior and the Principle of Least Effort”, 1949-B. Trnka’s work on English has the title “A Phonological Analysis of Present-Day Standard English”, 1935. Cf. Carl-Gustaf Söderberg, “A Typological Study on the Phonetic Structure of English Words with an Instrumental-Phonetic Excursus on English Stress” /Travaux de l’Institut de phonétique de Lund, publiés par Bertil Malmberg, I, 1959/. — Problems of linguistic statistics have been treated by Pierre Guiraud, particularly in “Les caractères statistiques du vocabulaire”, 1954, and “Problèmes et méthodes de la statistique linguistique”, 1960. F. W. Kading’s pioneer work had the title “Häufigkeitswörterbuch der deutschen Sprache”, 1898. Concerning Liberman’s opinion about the problem discussed in this chapter, it is expressed in his article in Sol Saporta’s “Psycholinguistics”, quoted above. — The general problem of the application and validity of statistics in linguistic description was taken up recently by Hans Karlgren in “Die Tragweite lexikalischer Statistik” /Sprâkvetenskapliga sällskapets i Uppsala förhand-lingar 1958–60/. — The reference to Herdan concerns “Language as Choice and Chance”, pp. 79–80.Google Scholar

Copyright information

© Springer-Verlag Berlin · Heidelberg 1967

Authors and Affiliations

  • Bertil Malmberg
    • 1
  1. 1.Faculty of ArtsUniversity of LundSweden

Personalised recommendations