Leonard Bloomfield (1887-1949)

Whereas the Linguistic Society of America celebrated its centennial in 2024, with the first issue of its journal “Language” appearing in 1925, it was in September 1926 (exactly a century ago) that the most ambitious early paper appeared: Leonard Bloomfield’s “A set of postulates for the science of language”.

For centuries, linguists had described the grammatical systems of other languages in terms of the categories and technical terms of Latin grammar – terms such as adverb, participle, infinitive, accusative case, plural number, or subjunctive mood were used widely and impressionistically. The broad strokes of the grammars of other languages could be characterized in this way, but since Boas (1911), it had become clear that on closer inspection, each language had its own system, and linguists often needed new concepts that did not exist in traditional European grammars.

Bloomfield had been trained as an Indo-Europeanist (also at Leipzig University, in 1913), but he also knew ancient Sanskrit grammar (cf. Rogers 1987), and he had studied Tagalog and a range of Indigenous languages of North America (especially Menomini, an Algonquian language). And the 1920s were a time of widespread conceptual innovation after the terrible Great War. But whatever led him to write his “postulates”, the paper was an amazing attempt to put linguistics on a new footing.

Robert A. Hall Jr writes in his biography of Bloomfield (1990: 29):

“Bloomfield’s concern with underlying method found its fullest and clearest expression in his ground-breaking and still fundamental article “A set of postulates for the science of language”, in volume 2 ofLanguage (1926). That it was reprinted in 1949 in the International Journal of American Linguistics in lieu of an obituary may be interpreted as an indication that many considered this to be Bloomfield’s legacy. For clarity, precision, concision, and true rigor (not rigidity!), this article has no equals. In it, Bloomfield sets forth the basic assumptions on which his treatment of language in his 1933 book was founded, and on which all sound discussion of linguistic matters has rested ever since (with, of course, some further improvements).”

Hall is perhaps exaggerating slightly, because other thinkers (structuralists, Chomsykans, and others) have contributed to our thinking about the foundations of grammar after 1926, but before Bloomfield’s paper, there was little that could be compared to it (maybe Paul’s 1880 Principien der Sprachgeschichte, though that was a much  less striking break with the tradition). Bloomfield recognizes that one cannot simply reject the traditional conceptual framework of Western linguistics (its traditional stereotypes) – one needs new foundations, which are independent of the patterns and concepts of Western grammar(s).

Moreover, one needs clear terminology, and rigorous definitions. Linguists of course know roughly what their terms mean, but traditionally, these terms have not been defined – linguists learned them from examples, and they accepted them from the tradition that sometimes goes back to late antiquity (or even to Aristotle).

Bloomfield’s immediate motivation was to avoid disputes with psychological notions that he thought had no (or unclear) relevance to linguistics – he wanted to empasize the autonomy of linguistics (as made particularly clear by his inaugural paper explaining the need for a special association of linguists; Bloomfield 1925).

But what’s impressive and unprecedented is his definitions of a range of basic technical terms, such as:

Definition 1: An act of speech is an utterance.

Definition 4: The totality of utterances that can be made in a speech-community is the language of that speech-community.

Definition 6: The vocal features common to same or partly same utterances are forms; the corresponding stimulus-reaction features are meanings.

Definition 9: A minimum form is a morpheme; its meaning a sememe.

Definition 10: A form which may be an utterance is free. A form which is not free is bound.

Definition 11: A minimum free form is a word.

Definition 12: A non-minimum free form is a phrase.

Definition 13: A bound form which is part of a word is a formative.

Definition 23: Recurrent sames of order are constructions; the corresponding stimulus-reaction features are constructional meanings.

What is striking is that these definitions are simple and form a coherent system – an idea that had not come up (at least not prominently) in linguistics before. In the 1926 paper, Bloomfield does not explain the definitions in detail, but many of the same ideas are spelled out in detail in his 1933 book Language, which was extremely influential in North America until the 1960s.

Rigorous terminology was not important in the dominant approaches of the 1960s and later decades, but over the last 15 years, I have come to think that Bloomfield was basically on the right track in the 1920s and 1930s: Without rigorous definitions of basic concepts of general grammar (= basic comparative concepts), linguists will hardly be able to understand what is general about grammars.

The latest fruit of my terminological efforts is called Grammaticon, a database of hundreds of comparative-concept terms: https://grammaticon.clld.org/. Personally, I was more influenced by Mel’čuk and Wierzbicka than by Bloomfield (as noted here), but without Bloomfield, many subsequent developments wouldn’t have been possible.

In my 2021 paper on the term affix, I made crucial use of the distinction between free forms and bound forms, following Bloomfield, who introduced these terms. However, I do not define affix as “a bound form which is part of a word” (cf. Bloomfield’s formative), because “word” is not a natural concept (cf. my 2023 definition).

My 2020 paper on the definition of the term morph revisits Bloomfield’s morpheme. This term existed before Bloomfield (it was coined by Baudouin de Courtenay in 1880), but he really made it well-known – but without providing a clear discussion of what its boundaries are (and resulting confusion in subsequent decades; see Mugdan 1986).

And finally, it appears that Bloomfield was also the first to define construction as a technical term of grammar (see my 2023 definition, which, however, makes no reference to Bloomfield).

In hindsight, Bloomfield’s approach seems quite prescient to me, though for a long time, I found definitions as boring as most of my colleagues. However, over four decades in the field, it has become clear to me that there is a lot of talking past each other in our field, and without well-defined  technical terms, we can hardly make progress. Retro-defining traditional terms such as word or serial verb construction  may not be very useful  in the end, but not having such definitions is much worse, because we may be led to chase after stereotypes.

Thus, I celebrate Bloomfield’s innovative approach to grammatical methodology and theorizing, which became public through his article in September 1926, exactly a century ago.

References

Bloomfield, Leonard. 1925. Why a linguistic society? Language 1(1). 1–5.

Bloomfield, Leonard. 1926. A set of postulates for the science of language. Language 2(3). 153–164. (doi:10.2307/408741)

Bloomfield, Leonard. 1933. Language. New York: Holt, Rinehard and Winston.

Boas, Franz. 1911. Introduction. In Boas, Franz (ed.), Handbook of American Indian languages (Volume 1) (Bureau of American Ethnology Bulletin 40–1), 5–83. Washington: Government Printing Office.

Hall, Robert A. Jr. 1990. A life for language: A biographical memoir of Leonard Bloomfield. Amsterdam: Benjamins.

Haspelmath, Martin. 2020. The morph as a minimal linguistic form. Morphology 30(2). 117–134. (doi:10.1007/s11525-020-09355-5)

Haspelmath, Martin. 2021. Bound forms, welded forms, and affixes:  Basic concepts for morphological comparison. Voprosy Jazykoznanija 2021(1). 7–28. (doi:10.31857/0373-658X.2021.1.7-28)

Haspelmath, Martin. 2023a. Defining the word. WORD 69(3). 283–297. (doi:10.1080/00437956.2023.2237272)

Haspelmath, Martin. 2023b. On what a construction is. Constructions 15(1). (doi:10.24338/cons-539)

Martin Haspelmath

View all posts

The text only may be used under licence Creative Commons Attribution 4.0 International. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.