Grammatical gender

Categories, inflection, congruence

In languages that have genera, except in a few special cases, each noun is clearly assigned a genus. This has the effect that other words referring to the noun are inflected depending on the genus of the noun, i.e. they change their form. The genus is the grammatical category underlying this inflection. For example, the adjectives in the expression

to the genus (masculine) of the noun prince. The rule of congruence exists between the noun and its attributes and the article one (in the example we have two separate attributes, hence inflection twice). Such an adoption of a grammatical category in an inflection is called congruence. The noun prince, on the other hand, carries the genus "masculine" as a fixed feature, it does not arise there by congruence with the grammatical environment.

The genus of a noun is therefore firmly assigned to it, so that there is no inflection of nouns according to the genus. Cases like Koch / Köchin (so-called movierung) are not a counterexample, because they are derivations of a new word, not different inflections of the same word. This can be seen from the fact that the base of the derivation, namely cook, itself already carries a masculine gender, the derived cook is a new word, with a different gender. Hence the difference that the plural of a noun is very much an inflectional form, as opposed to the genus: the plural only comes to the root of the noun through the inflectional form, this feature is not already present in the root of the word.

Thus, to prove whether a language has genus, it is important not simply to look at the "gender" of a noun, but genus as a grammatical feature only shows up in the inflection of other words due to congruence rules. Which parts of speech in a language are congruent with the noun in terms of genus varies from language to language (see section on genus congruence). In some cases, a pronoun may also have its own genus, with which other words then congruence (see section Genus of pronouns).

In some languages, one can infer the noun's genus from its form and from its inflections by numerus and case. Or there is a connection between word meaning and genus. Such morphological (concerning the forms of the word) or semantic (concerning the meaning) connections, however, do not constitute the genus; this is rather characterized by the genus congruence of other words. Distinguishing this is important because such contexts often do not cover all nouns or have some exceptions, whereas congruences are unambiguously determined by the genus, even if the assignment of the genus to the noun seems irregular. For example, in German words denoting only female persons or animals regularly - but not invariably - have feminine genus, while diminutives on "-chen" have neutral. The word "Mädchen", for which these two rules contradict each other, nevertheless has an unambiguous genus that unambiguously determines all congruences in the same sentence, namely the neutral one. In the subordinate clause "the girl who wore her hair down" there can be no feminine congruences with "girl". - For the distinction of genera from morphological and semantic properties of nouns, see the section Dependencies of the genus.

Other categories of the noun

Besides the genus, there are other categories of the noun that can also affect congruent other words. Here we are concerned with the interaction with the genus. Often, genus, numerus and - if present in the language - case are mentioned as the three categories responsible for noun congruences (KNG congruence), but other categories, especially definiteness and vivification, can also play a role.

Numerus

Some languages distinguish the genera only in the singular, not in the plural. This is true for German and the other Germanic languages except for Faroese and Icelandic, as well as for the Northeast Slavic languages Belarusian, Russian and Ukrainian and the Southeast Slavic languages Macedonian and Bulgarian. In these languages, in the plural, the congruent articles and adjectives in all cases are independent of the genus of the noun:

• old gentlemen - plural without article: N/A: old gentlemen, G: old gentlemen, D: old gentlemen - plural with article: the / the / the old gentlemen
• old lady - plural without article: N/A: old ladies, G: old ladies, D: old ladies - plural with article: the / the / the old ladies

All Baltic and many Romance languages, on the other hand, distinguish two genera in the plural form, namely masculine and feminine, as do the Semitic languages.

When the genus distinctions in the plural have disappeared, the underlying genus of words without singular (the plural nouns) cannot be determined from the congruences, but at best from the word history, as in the case of Kosten and Ferien, which derive from the Middle High German feminine koste and from the Latin feminine plural word feriae. In dictionaries, then, "plural" is often given instead of a genus.

In some daughter languages of Latin there are so-called ambigenous nouns, which continue a remnant of the old class of neuters. These nouns always behave like masculines in the singular, but like feminines in the plural. In French and Italian, these are only a handful of words, while in Romanian this scheme has covered a large number of nouns (several thousand); the group of these nouns is therefore often labelled as neuters in Romanian, although it has no forms of its own, but merely makes use of the respective forms of the other two genera, depending on the numerus. Albanian also has ambigenera.

Examples of Ambigenera:

• in Italian: il labbro (sg.m.def.) - le labbra (pl.f.def.), the lip - the lips
• in French: l'amour mort - les amours mortes (pl.f.def.), the dead love - die toten Liebensowie le vieil homme (Sg.m.def.) - les vieilles gens (pl.f.def., only before the subst.), the old man - die alten Menschen
• in Romanian: scaunul (sg.m.def.) - scaunele (pl.f.def.), the chair - the chairs

If one does not presuppose that genus and numerus are completely independent of each other, one arrives at a description in which the usually two numeri each break down into one or more classes and for each word it is determined in which singular and plural class it lies, if it occurs at all in the respective numerus. In this model, for example, there would be four classes for German: the three singular genera and a common class for all plural words including the plural nouns, since the congruences of plural words do not depend on the genus of the singular word. In Romanian or French there would also be four classes, namely masculine and feminine each in singular and plural, and the ambigenera also fit into the scheme. For the nominal classes of the Bantu languages this is done; for the genera of the Indo-European and Semitic languages it is not usual.

Case

Kasus is a category that changes the word form on the noun and on congruent words - more or less the same ones that also congruent with the noun in terms of genus. In German, nouns are hardly changed at all (only genitive-s for masculine and neutral words in the singular and dative-n in the plural), so that the case is mainly shown at the article as well as then at the adjective if the article is missing or has no genus- and case-specific ending. Fully developed cases exist in most Slavic, Baltic and Insular Norse languages, whereas the other Germanic and Romance languages have abandoned the case distinction.

Definiteness

The definiteness of a noun is a grammatical category used to indicate whether the noun means definite things or persons or indefinite ones. It can act together with the categories of genus, numerus and case on congruent words as described below.

In German and many other languages definiteness is expressed by using the definite article, which is congruent with the noun by genus, numerus and case. The substitution of the indefinite article for the definite article, however, is not taken as an inflection of the article according to the category of definiteness. The article may also be attached to the noun itself as a suffix (so in Scandinavian languages depending on the context, and in Romanian and Albanian) or as a prefix (so in Arabic and Hebrew). Other congruent words such as adjectives are less often affected. For example, in Hebrew not only the noun is affixed with the genus- and numerus-independent article prefix ha-, but also adjectives, and the construction of the whole nominal phrase depends on genus and definiteness:

• schloscha jeladim tovim (indef., mas.) - three (basic form schalosch) good (GF tov) children (GF jéled)
• schlóschet hajeladim hatovim (def., mask.) - the three good children
• shalosh jeladot tovot (indef., fem.) - three good girls (GF jalda)
• shalosh hajeladot hatovot haélle (def., fem.) - these (genusunabh. GF élle) three good girls

liveliness

In many languages a distinction is made in grammar between animate and inanimate nouns, the boundary being usually between man and animals on the one hand, and plants, things, and abstracts on the other, and sometimes between man and animals. In the Anatolian languages, an extinct branch of the Indo-European languages, animate is the main criterion for assigning nouns to the two genera.

In languages with a different system of enjoyment, the animate category can further differentiate the genera. Examples:

• In some languages whose genus system no longer has any reference to natural genders, different pronouns are nevertheless used depending on the gender of persons, for example in Danish, where the personal pronouns of the 3rd person singular for things are genus-dependent den (utrum) and det (neuter), but for persons gender-dependent han (masculine) and hun (feminine). If we regard this as a genus difference, there are four genera instead of two.
• Similarly, in English, he, she, and it (with the possessive pronouns his, her, and its) are distinguished mainly by vivification and natural gender, although there are otherwise no genera.
• In some Slavic languages, in the singular of masculine words and in the plural, the accusative has the same form as the genitive for animate, and as the nominative for inanimate. They then distinguish between an animate masculine for persons with masculine sex (Czech nový král = new king, genitive nového krále, accusative nového krále) and an inanimate masculine (Czech nový hrad = new castle, genitive nového hradu, accusative nový hrad), so in a sense they have four genera.
• In Swahili there is a pair of classes (class 1/2 for singular/plural) exclusively for animate beings, but animate beings can also be in other classes. They then have part of their congruences according to their class and part according to class 1/2 because of the property of being animate. Again, this increases the number of genera if you consider it a genus difference.

Grammars refrain from considering these distinctions as separate genera and instead describe the variations in animate nouns as complementary rules about declension and genus congruence.

In German, animacy - as in the first two examples above - mainly concerns pronouns (wer/was, jemand/etwas); see the section Genus von Pronomen.

Vividness is given by the meaning of the word, so that synonyms have the same vividness. With the genus, on the other hand, it happens that it can change when a word is replaced by a synonym: "a man and his profession", but "a person and his profession", or "a woman and her profession". If such a thing does not occur anywhere in the language, it is questionable whether it is a genus distinction.

An example of another categorization by word meaning that is not usually considered a genus are the count words in East Asian languages, which can be considered modifications of the numeral word or demonstrative pronoun that precedes it and depends on the noun that follows it.

Inconclusive genus

The genus of a noun is not always unambiguous, even if it is not a matter of accidental sameness (homonymy) of different words, as in der / die Kiefer or der / das Tau. Sometimes the same word is used regionally or individually with different genus, without one of the genera being considered correct and the other incorrect: der / das Gummi, der / das Katheder, der / das Cola, der / die Abscheu, der / die Dispens, der / das Traktat. In some words, the differentiation of the meanings of the same word (polysemy) has simultaneously differentiated the genus: the / the shield, the / the merit, the / the corpus, the / the recognition, the / the part, the / the lake.

On the ambiguity of the genus in plural nouns and ambigenera, see the Numerus section.

Some apparent ambiguities of the genus also come from the expectation that the genus must always correspond to the biological sex. Thus, the ancient grammarian Dionysios Thrax (2nd century B.C.), in his Greek Grammar, mentions two more that "some add" to the usual three genera that he regards as undoubtedly existing:

• Γένος κοινόν (génos koinón "common gender"; Latin genus commune) denotes the genus expression of nouns that are used as masculine or feminine depending on the biological sex of the designated being (for example, in Dionysios (ho / hē) híppos (horse)). In German such words are rare (der / die Präses, der / die Hindu, der / die Azubi), in French they are frequent (un / une enfant, le / la ministre, le / la pianiste and other personal names on -e). They behave like two polysemous words with different genus.

• Γένος ἐπίκοινον (génos epíkoinon "blended gender"; Latin genus promiscuum or genus epicoenum) refers to the genus expression of nouns with a clearly fixed genus whose meaning includes beings of both biological sexes. As examples Dionysius gives (hē) chelidōn (swallow) and (ho) aetós (eagle), i.e. a feminine and a masculine for animals, where there are no specific words for males and females, but he does not mention neuters or personal nouns.

The Latin designations were coined by Aelius Donatus (4th century AD), who adopted the classification of Dionysius with modifications. The German designations from early New High German translations of Donatus are hardly used today. In the general German vocabulary, the word epicönum (also epikoinon) is found for a noun belonging to the genus epicoenum.

These terms are not always used consistently. In English and French, the adjective epicene or épicène is used in both meanings described above. Genus commune is also used synonymously with utrum.

Nominal class

The term nominal class was introduced in the 19th century, initially with reference to a classification of nouns in Bantu languages (such as Swahili). As with genus, which was known from ancient Greek and Latin, the noun serves as the reference point of other words in the sentence that are congruent with it; nominal class thus satisfies the same definition as genus. One usually speaks of genus when dealing with classical languages such as Sanskrit, Hebrew, Greek, and Latin, and with other Indo-European and Semitic languages: these have two or three genera, usually one of which is masculine and one feminine. In languages with finer class divisions, and when comparing quite different classification systems, one tends to speak of (nominal) classes, but genus is also used in this way. It is a rather historical distinction without a sharp dividing line.

The nominal classes of the Bantu languages differ from the genera of the Indo-European and Semitic languages in the following respects:

• Singular and plural are counted separately (see the Numerus section).
• One can take the occurring pairings of singular and plural class as genera or as declension classes, since the congruences result from the form of the noun in singular and plural, except for the peculiarities described in the section animacy.

gender of pronouns

This section describes the situation in German. In other languages with a genus, other rules may apply.

In the narrower, more modern sense, pronouns are words that take the place of a noun or proper noun in a sentence: Personal, indefinite, and interrogative pronouns, also possessive and demonstrative pronouns when they do not accompany a noun. They have a genus that can be taken up by a possessive pronoun (genus of possessor, not possessed) or a relative pronoun. Examples:

1. He, putting on his jacket, ...
2. She, putting on her jacket, ...
3. It, putting on its jacket, ...
4. Mine (instead of: my car), which does its service, ...
5. This (instead of: this machine), which has earned its operating costs, ...
6. Someone who puts on his jacket...
7. Who is it that puts on his jacket?
8. You can't always hide your feelings that suddenly come over you.
9. Something that doesn't do its job...
10. What is it that is not doing its job?

If they are personal pronouns (Ex. 1 to 3), the noun they represent has been mentioned before, and they almost always take its genus (for counterexamples, see below). Possessive pronouns and demonstrative pronouns that do not accompany a noun (Exs. 4 and 5) have the genus of the missing noun, which can be ambiguous (in Ex. 4 mine for my car or mine for my car). Indefinite and interrogative pronouns (Ex. 6 to 10) have their own genus, which they have not taken from a noun or proper name. They are masculine for persons (someone, man, who) and neutral for things (something, what), regardless of the genus and also the sexus of what is meant. If one wants to incorporate the natural gender for persons with the help of the genus, one can say something like:

• One who puts on his jacket...
• One who puts on her jacket, ...

Indefinite pronouns followed by relative pronouns are often replaced by interrogative pronouns without changing the genus:

• He who is late is punished by life. (instead of: Someone who is late is punished by life. )
• What doesn't interest me, I don't read. (instead of: Something that doesn't interest me, I don't read. )

Constructio ad sensum

Personal pronouns do not always take the genus and numerus of the noun they represent. Especially for a person whose natural gender is known, people often use pronouns with the corresponding genus. The pronoun is then understood to refer less to a word than to the person it denotes. Similarly, singulars denoting a plural of things or persons may be resumed by plural pronouns, and vice versa. Such a choice of pronoun is called Constructio ad sensum. As with all stylistic devices that violate formal grammar, their use is controversial. Often the break is mitigated by interposing another noun or a name with the new genus as the predicate noun:

• The girl who just came in is called Susanne. She works here.
• The other hostage was a man. He was about forty years old.

In general, however, personal pronouns cannot refer to predicate nouns:

• The buttercup is a meadow flower. It (not: she) blooms yellow.

Reflexive possessive pronouns (i.e. his/her, so that his/her own is meant) should follow the genus of the reference word, be it a noun or pronoun, if possible:

• The girl had tied her (not: her) hair into a ponytail.
• A girl had come in. She had tied her (not: his) hair into a ponytail.
• A girl had come in. She had tied her (not her) hair into a ponytail.

Relative pronouns always follow the genus of the word to which they refer:

• The student is a nice girl who (not: who) is also very hardworking.

The genus is a fixed grammatical category of the noun, which can be marked on the noun itself. With Italian nouns such as origano "oregano" or salvia "sage", one can usually recognize the masculine or feminine genus by the ending (-o or -a); with German nouns such as sage, columbine, einlei, one cannot recognize it. However, this is not decisive; what is important is that the genus is marked on other words in the sentence that are congruent with the reference noun, i.e. have the same genus. For example, in German the adjective adjective congrues with the reference noun in the genus: fresh sage - fresh parsley - fresh basil. Frequent is the genus congruence of determiners and attributes of a noun. With participles in the formation of certain tenses, as in Russian and Arabic, or in the passive voice, in numerous languages parts of the predicate also congruent with its subject in genus and not only in numerus. In Romance languages, the same participle congrues with the subject in passive formations, but not in perfect formations.

Demonstrative pronouns can also be congruent with their predicate noun in subject function, as in Latin and Italian (faccenda is feminine, problema masculine):

Questa è una faccenda seria - This is a serious matter

Questo è un problema serio - This is a serious problem

Congruence of the article

For German, it is common to indicate the genus of a noun by giving the form of the definite article to it. However, this is not possible for all languages with genera:

• Many languages, including Latin and Russian, do not have articles or any other marker of the definiteness of a noun.
• To the extent that there is such marking, it may be independent of genus and numerus, as in Hebrew and Arabic, and it may be on the noun itself, as in Scandinavian and Semitic languages, that is, not on any other word in the sentence.

Congruence of the adjective

Adjectives usually change their form according to the genus, numerus, and - if present in the language - case of the associated noun. In German, moreover, even with the same genus, case, and numerus, there are up to three forms, depending on whether the adjective is used attributively or predicatively, and in the former case whether it is preceded by a definite article or demonstrative pronoun. Similar distinctions exist in other languages; here are examples from Danish, German, and Russian:

Abbreviations:

m = masculinef =
feminineumn =
neutrumu
= utrum
p = plural (in these three languages only one plural for all genera)

Congruence of the number word

In some languages, numeral words also differ beyond the one for nouns of different genera, as in Hebrew or in Bantu languages such as Swahili. In Russian, the numeral words are declined, but only the two differs by genus. In Hebrew there is the peculiarity that the numeral words for counting masculine objects have feminine endings and vice versa.

Congruence of pronouns

Personal and demonstrative pronouns in languages with genera usually depend on the genus of the denoted noun. Possessive pronouns can depend both on the genus of the possessor (his / her) and on that of the possessed (his / his). In many languages, only one of the distinctions is made.

As function words, pronouns do not necessarily have to be words in their own right; they can also take the form of clitics or affixes. Example: The phrase ihn / sie sehen is called le / la voir in French and verlo / verla in Spanish. In German, they are two words that also occur separately - for example, as an answer to a question - and between which one can insert other words as desired; in French, le / la does not occur separately as an object pronoun, but only as an unstressed clitic before the verb, and in Spanish it is written together with the verb as a suffix, which is less a linguistic difference than a purely orthographic one.

In Swahili, personal pronouns appear only exceptionally as independent words, mainly to underline the grammaticalperson and therefore genus-independent. Otherwise, their function is taken over by genus-dependent verb prefixes, such as amelitazama (he / she looked at it; with the subject prefix a-, the object prefix li- and in between the tense prefix me- for the perfect). Unlike in the previous examples, these pronominal prefixes not only replace subject and object, but at the same time serve to conjugate the verb, which is not changed at the verb stem: mama amelitazama gari (the woman looked at the car; literally woman she-has-looked-at-it car). So here the verb form shows congruence with the genus of subject and object - unless one considers the prefixes as clitics and their writing together with the verb stem only as an orthographic convention.

Possessive pronouns also have the form of affixes in some languages, which then congruent with the genus of the possessor, for example, in Hebrew sefer / sifro / sifrah (book / his book / her book), sfarim / sfaraw / sfarejha (books / his books / her books).

Congruence of the verb

With regard to the possible genus congruences, finite and infinite verb forms behave differently. Finite verb forms are those on which a variety of grammatical categories such as person, numerus, tense, genus verbi and mode are marked. In German and other Indo-European languages, the numerus of the subject is marked on the finite verb, but its genus is not; moreover, in other languages, the genus and numerus of the subject and object may also be marked on the verb. An example of this from Swahili, a language with agglutinative formed verbs, was discussed in the previous section.

However, such congruences also exist in finite verb forms of inflectional languages. An example of subject congruence from modern Hebrew:

• Shmuel raqad. Atta raqadta. Lea raqda. Att raqadet. (Shmuel danced. You^{(m) danced}. Lea danced. You^{(f) danced}.)
• Shmuel yirqod. Atta tirqod. Lea tirqod. Att tirqedi. (Shmuel will dance. You^(m) will dance. Lea will dance. You^(f) will dance.)

In Biblical Hebrew there are exactly the same forms with different word order and slightly different meaning; but the subject congruence is the same. The object congruence is limited to Biblical Hebrew when the object is a pronoun:

• ta'asvennu (you^(m) will leave him)
• ta'asveha (you^(m) will leave them)

Infinite verb forms are used in the sentence in a similar way to other parts of speech, namely infinitives like nouns and participles like adjectives or adverbs. In terms of genus congruences, they inherit the properties of these parts of speech. For example, German present participles have the property of adjectives to be genus-congruent with the noun when used attributively: a smiling salesman, but a smiling saleswoman. The following French examples show that in this case, genus congruences can occur with both the subject and the object of the verb:

• les mots^(m) qui étaient dits (the words that were said)
• les paroles^(f) qui étaient dites (the words that were said)
• les mots qu'il avait dits (the words he had said)
• les paroles qu'il avait dites (the words he had said)

An intermediate position between finite and infinite verb forms is occupied by those forms which are participles in linguistic history, but next to which there is no finite verb in the same sentence, namely when a verb to be is not explicitly added because it is optional in the language.

• Hebrew: Shmuel roqed. Atta roqed. Lea roqedet. Att roqedet. (Shmuel dances. You^{(m) are} dancing. Lea is dancing. You^{(f) are dancing}. Actually: Shmuel [is] dancing. ... You^(f) [are] dancing.)
• Boris tanzewal. Ty tanzewal. Anna tanzewala. Ty tanzewala. (Boris danced. You^{(m) danced}. Anna danced. You^{(f) danced}. Actually: Boris [is] danced having. ... You^(f) [are] having danced.)

Such forms are perceived by speakers as finite verb forms. In this way, at the expense of person congruence, genus congruence with the subject can come about, even if it does not otherwise occur in the language.