From Grammatical to Natural Gender

by Jesse Archibald-Barber

copyright 2001

Gender can be a complicated category of language, and language change. To help clarify the issue, it is important to distinguish two types of gender systems, one according to grammatical conventions, the other according to natural conventions. The traditional theory holds that at one time English had a grammatical gender system, but made the transition to a natural gender system “in the East Midlands of England by the early twelfth century” (Smith 130). However, recent scholarship by Hans Platzer critiques the traditional theory, revealing that the issue is much more complex. Rather than a systematic transition from a homogeneous grammatical gender system in OE to a natural gender system by ME, Platzer reveals that the history of English gender marking has always been characterized with “conflicting tendencies” (36). The acute insights of Platzer’s research will be discussed below, but first a definition of grammatical and natural gender and their importance to the history and structure of the English language will be given.

A grammatical gender system uses inflections to indicate whether a referent’s gender is masculine, feminine or neuter. Inflections are affix-tags attached to words, and involve a more complex system of declensions for nouns, pronouns, adjectives and determiners, which must agree, or concord, with the noun’s gender. For instance, if a noun is grammatically masculine, such as cyning (king), it requires a corresponding masculine inflection, and any determiner, pronoun or adjective related to the masculine noun must also take the appropriate declension. Hence, se cwic cyning (the living king) is grammatically correct, whereas seo cwicu cyning is incorrect because seo is a feminine determiner and the –u on cwic is a feminine inflection for the adjective. Rather, seo cwicu cwen (the living queen) is grammatically correct. It should be noted that in the above examples, the grammatical gender of the nouns happens to correspond to the natural gender of their referents -- male-king, female-queen. However, a grammatical gender system can also run counter to a referent’s biological sex. A classic instance is se wifmann (the woman). Although the referent’s natural gender is feminine, because gender is grammatically conveyed, the noun happens to be masculine, and is therefore inflected according to a masculine declension, taking a masculine form of the determiner as well. Any other word related to wifmann must also inflect itself as masculine for purposes of grammatical agreement, or concord.

A natural gender system, on the other hand, indicates gender according to its referent’s biological sex. In a sense, in a natural gender system, gender is hardly even a category (Classen 101), for nouns and pronouns reveal the referent’s gender themselves, and determiners and adjectives take the same form, whether modifying a masculine, feminine or neuter referent -- hence, the happy woman, the happy man, the happy cat – she, he, it is happy. It must be further noted, however, that neither gender system is less ambiguous than the other, and neither is universally used (Classen 98). Nonetheless, as the simplicity of a natural gender system works well for PDE, one may ask, why did grammatical gender arise in the first place? And if the grammatical gender system once worked, why was there a subsequent shift to a natural gender system?

Traditional theories attribute the origin of grammatical gender systems to the human-primitive tendency to anthropomorphize nature, when one endows human qualities to non-human creatures and objects (Ibrahim 16). Another theory raised by James Frazier of Golden Bough fame is that early men and women spoke a differently inflected form of the language from one another (Ibrahim 19). However, these kinds of speculations assume a universal tendency in human linguistic relations to the world, and break down with the fact that many languages have never had a grammatical gender system (Ibrahim 24). Rather, Ibrahim argues that grammatical gender is an “accidental outcome of the linguistic development of some languages” (102). Indeed, studying the human-animate nouns of OE, one finds that nearly all of their grammatical genders correspond to their natural genders to begin with (Platzer 35). Instances like se wifmann are exceptions and not the rule, and even then such words often took the natural gender rather than their original grammatical gender (Platzer 42).

As such, there is a general consensus that “at some stage in its development, [grammatical gender] must have been an extension of natural gender into the sphere of language” (Ibrahim 30), allowing for a more articulate distinction of gender in human-animate nouns. The main function and advantage of grammatical gender, therefore, is its ability to clarify syntactic agreement in otherwise ambiguous cases (Ibrahim 26). Thus, a language based on grammatical gender can indicate the gender of an unsexed noun with inflections, whereas the English natural gender system must add the words ‘male’ or ‘female’ to make the referent’s sex clear, such as ‘a male Canadian’ instead of ‘a Canadianat.’ But if grammatical gender is so effective as a linguistic marker, the question still remains, why did English change to a natural gender system?

To fully answer this question, one must follow Platzer in dispelling traditional misconceptions about the status of the grammatical gender system in OE. The traditional theory is that in OE it became very difficult to distinguish gender, due to the leveling of determiners to one form (i.e. se, seo, paet to ‘the, that), and the general loss of inflectional endings (Classen 97). Such an explanation, however, begs a further question: why did English lose its inflections in the first place, when other European languages clearly function fine with them? Amid vast speculation, two related reasons stand out, one linguistic, and the other social-historical. As David Crystal explains, linguistically, “the most obvious explanation is that it became increasingly difficult to hear [inflections]” (32) because most words had the stress at the beginning. This initial stress “readily g[a]ve rise to an auditory problem at the end [, . . .] especially when there were several endings which were phonetically very similar, such as –en, -on, and –an” (32). However, one may still ask, why did English lose its inflections and not other Germanic languages, which have a similar stress pattern?

The primary explanation is that, in addition to linguistic-phonetic ambiguity, social-historical conditions in the late OE period facilitated the loss of inflections. Baugh argues that the Norman Conquest “brought about conditions favorable to such changes” (200). By making French the language of prestige, and “English the language mainly of uneducated people[,] the Norman Conquest made it easier for grammatical changes to go forward unchecked” (200). As English writing institutions were supplanted, grammatical features altered according to speech patterns that may have otherwise been maintained by clerical custodians of the language. Hence, with the loss of inflections and the further leveling of determiner-forms unmarked for gender, English came to rely on the referent’s natural sex to indicate gender, implying that the loss of grammatical gender was already underway in spoken OE.

However, although this traditional theory makes sense, many contemporary scholars believe that more personal and psychological factors were involved. The above reasons of inflectional loss and determiner leveling were no doubt influential factors and products of the transition from grammatical to natural gender systems, but many scholars feel they are not the main reasons. For instance, Classen contests the theory “that natural gender sets in after the confusion arising from the loss of inflections” (98). Such a theory treats natural gender as a substitute for a lost grammatical system, when, as mentioned above, grammatical gender is an extension of natural gender. Instead, Classen argues that “the evidence which is available goes to show that natural gender came in by way of the personal pronouns” (100), positing that OE speakers made a strong distinction between human and non-human categories, including sexless and non-living things (101). This is not to say that personal pronouns did not exist in OE, but that as OE gender pronoun distinctions began to level, the advent of ‘he, she, and it’ augmented the tendency to distinguish between human and non-human categories. As such, in addition to linguistic tendencies, and social conditions, Classen believes that psychological choices played the key role in the transition from grammatical to natural gender system.

Platzer takes Classen’s critique a step further, arguing that the notion that OE ever had a homogenous grammatical gender system, which became natural after the loss of the inflectional system, is simplistic and deceiving. Focusing on human animate nouns, he argues that “in contradistinction to the rest of the system, human animates show a marked tendency towards natural gender assignment” (35). What Platzer means is that OE had neither a totally systematic grammatical or natural gender system, but that the presence of both created “conflicting tendencies” (36). Indeed, as a result of the conflicting tendencies, Platzer argues that even the natural gender system “does not equate the gender of the noun with the sex of its referent. Rather, gender is merely related to the class of referent involved so that human animates take masculine or feminine gender while all other classes of referents (animals, plants, objects, abstracts) receive neuter gender” (36). Therefore, the English gender system in general is not so much based on natural sex, or grammar, but on the class distinction between (human) animate and (non-human) non-animate referents.

For instance, Platzer cites statistical evidence that in the class of human animates, grammatical gender already coincided with natural gender in over 90% of the lexical types involved (38). Moreover, as Blake points out, “even in Old English, grammatical gender had not been consistent, especially when a pronoun had been separated from the noun it referred to” (147). Natural gender is therefore the rule in OE human animates, while grammatical gender is the exception (Platzer 39). However, Platzer goes on to point out that gender marking is still not clear-cut in the case of non-animates. For one, the class of animates includes only human animates “despite the fact that [. . .] plants and animals are clearly animates as well” (36). Moreover, where one would assume that non-animates should take the neuter as their natural gender, the opposite was in fact the case. In the case of non-living things, “gender assignment in non-animates shows an active tendency away from natural gender” (40). For instance, rather that take the neuter gender, se stan (the stone) is masculine, and seo duru (the door) is feminine. This contradictory tendency created the “identical effect” (40) of weakening the neuter in both classes. The elimination of the neuter from human animates is clear enough, but in the case of non-animates, because they tended toward grammatical gender, they also avoided their natural gender of neuter.

The issue becomes more complicated in the case of animals, as they tended to “shift between the two classes of animate and non-animate depending on ‘familiarity or involvement’ (Biber 1999:317)” (36). In the case of animals, therefore, the attribution of gender depended more on pragmatics -- the attitude of “the speaker rather than on the referent” (36). Hence, one most likely uses ‘it’ when referring to a creature like an ant, but ‘he’ or ‘she’ when referring to a pet. In many cases, moreover, the gender of an animal, especially mammals and birds, was often indicated by specific lexical types for male and female referents, hence cow / bull, doe / buck. In general, however, as Baugh argues, the use of masculine and feminine gender for non-animates is not a function of grammatical or natural gender, but of attributive gender, a type of “personification and a matter of rhetoric, not grammar” (11). Nonetheless, Platzer succinctly points out that the conflicting tendencies of human animates toward natural gender, non-animates toward grammatical gender, and animals to shift between classes, leads to a difficult paradox: “The obvious trend towards natural gender in the animate nouns can only be fully realized by the total loss of the neuters from this sub system. However, as soon as this weakening of neuters is extended to the whole system, i.e. to non-animates as well, it results in a reciprocal strengthening of grammatical gender in the subset of non-animates” (44).

So, what finally pushed the non-animates into the natural gender system? Unfortunately, most current theories do not have a developed answer for this. One can only speculate that once the neuters were completely marginalized, the trend to keep the human animates separate from non-animates eventually collided with the trend to level determiners. As mentioned above, with the “gradual loss of pe and the replacement of the paradigm se, seo, paet by indeclinable that” (Fischer 296), it no longer became possible, or necessary, to establish grammatical agreement for inflected nouns. Furthermore, as non-animates increasingly became excluded from the human animate use of masculine and feminine categories, they were eventually leveled to the neuter. Overall, the move from grammatical to natural gender involves diverse linguistic, social and psychological factors that still require active speculation and research.

For Further Reading

Baugh, Albert C. A History of the English Language. 2^nd Edition. New York: Appleton-Century-Crofts, Inc., 1957.

Biber, Douglas. Longman Grammar of Spoken and Written English. Harlow: Longman, 1999.

Blake, N.F. A History of the English Language. Basingstoke: Macmillan, 1996.

Classen, E. “On the Origin of Natural Gender in Middle English” Modern Language Review. 1919, 14: 97-103.

Crystal, David. The Cambridge Encyclopaedia of the English Language. Cambridge: Cambridge UP, 1995.

Fischer, Olga. “Syntax.” The Cambridge History of the English Language. Vol. II. Ed. Norman Blake. Cambridge: Cambridge UP, 1992.

Ibrahim, Muhammad Hasan. Grammatical Gender: Its Origin and Development. The Hague: Mouton and Co., 1973.

Jones, Charles. “The Functional Motivation of Linguistic Change: A Study in the Development of the Grammatical Category of Gender in the Late Old English

Period.” English Studies. 1967a. 48: 97-111.

--- “The Grammatical Category of Gender in Early Middle English.” English Studies. 1967b. 48: 289-305.

Lass, Roger. “Phonology and Morphology.” The Cambridge History of the EnglishLanguage. Vol. II. Ed. Norman Blake. Cambridge: Cambridge UP, 1992.

Moore, Samuel. “Grammatical and Natural Gender in Middle English.” PMLA. 1921. XXXVI(1): 79-103.

Platzer, Hans. “‘No Sex, Please, We’re Anglo-Saxon?’ On Grammatical Gender in Old English.” View[z]: Vienna English Working Papers. 2001, 10: 34-47.

Smith, Jeremy. An Historical Study of English: Function, Form and Change. London: Routledge, 1996.

Traugott, Elizabeth Closs. “Syntax.” The Cambridge History of the English Language. Vol.1. The Beginnings to 1066. Ed. Richard Hogg. Cambridge: Cambridge

UP, 1992. (168-289).