CHAPTER 2

First Declension Nouns, Prepositions

The word “noun” is derived from the Latin noun nomen, “name”.

The principles outlined here for the declension of nouns apply also to adjectives (see Chapters 6 and 9) and pronouns (see Chapters 17 and 18).

The word “declension” is derived from the Latin verb declino, declinare, declinavi, declinatum (1) “to turn aside”, in the sense of varying a word’s form according to its grammatical function. This usage is much like that of inflectere “bend” as a general way of referring to morphological changes, i.e. “inflection”, or casus, a “falling” from the basic nominative case. Convention dictates that we should refer to the declension of nouns (and pronouns and adjectives), but to the conjugation of verbs. The difference in terminology has no practical significance.

neuter is an adjective meaning “neither”, in the context of grammar specifically neither masculine nor feminine. The dependence of the gender’s name on that of the other two reflects its comparative lack of importance; there are many fewer neuter nouns than masculine or feminine and, with the exception of Romanian, the Romance languages hardly use the neuter gender at all, generally subsuming it into the masculine. Already in classical Latin, masculine and neuter nouns, pronouns and adjectives are strongly akin, having the same endings more often than not.

English once used all these three genders, but the system is now largely obsolete, except with pronouns and pronominal adjectives, such as “he”, “she”, “it”, “his”, “her”, “its”. Of course, gray areas remain, as in “The Titanic was a great ship, but she did not survive her maiden voyage”. Natural gender and grammatical gender are interconnected in Latin, but there are languages in which natural gender has no role; Swahili, for instance, has eight gender classes, none of them masculine/feminine: one for animals, one for human beings, one for abstract nouns, one for diminutives etc.

Some principles of gender will soon become clear. For example, there are no neuter nouns in the first or fifth declension, very few feminine nouns in the second or fourth, very few masculine nouns in first or fifth: since the declension to which a noun belongs will usually be obvious (from the ending attached to the basic stem), knowing the declension willl therefore often help very substantially in determining its gender.

Is it easier to come to the question of gender with few preconceptions, as does the speaker of English nowadays, or from the perspective of a language which itself employs all three genders, such as German or Old English? The German words for “knife”, “fork” and “spoon” are neuter, feminine and masculine respectively, leading perhaps to a philosophically resigned acceptance of the challenge of determining gender in other languages.

In his Excerptiones de Arte Grammatica Anglice “Excerpts on the Art of Grammar in English”, one of the best known of the medieval vernacular grammars, written about 1000, Aelfric, abbot of Eynsham, felt it necessary to warn his readers that they must bear in mind that “nouns are often of one gender in Latin and another in English”. Speakers of Romance languages have a distinct advantage, since most nouns retain their Latin gender, with the neuter, as noted above, generally being subsumed into the masculine.

The exceptions, of course, will trap unwary speakers of Romance languages: flowers are masculine in Latin (flores), but feminine in French (fleurs), trees are feminine in Latin (arbores), but masculine in Spanish (arboles), leaves, which are neuter in Latin (folia), are feminine, not masculine, in Spanish (hojas) and French (feuilles).

As noted above, the technical term case is derived from the verb cadere, “to fall”, i.e. to inflect away from the basic (nominative) form. Latin originally had eight cases, presumably all distinct. The ablative is a conflation of three cases; see below. The development towards the Romance languages, which, with the exception of Romanian, have all but dispensed with inflection, was well under way by the classical period. This loss of inflection is most dramatically apparent in the singular of fourth declension neuter nouns, which show inflectional variation only in the genitive (see Chapter 11).

• Nominative, like “noun”, is derived from the noun nomen, “name”.

• Genitive is derived from the adjective genetivus, “by birth”. Note the change of spelling: “genetive” would be more correct, and only usage has made it wrong.

• Dative is derived from the verb dare, “to give”.

• Accusative is derived, by a strange error on the part of Marcus Terentius Varro, the best of the Roman grammarians, from the verb accusare, “to accuse”, under the influence of an equivalent Greek adjective which means both “accused” and “affected”. It is perhaps even stranger that later grammarians allowed Varro’s error to go uncorrected.

• Ablative is derived from the participle (see Chapter 19) ablatus, “taken away”, a derivation with little or no relation to many of the case’s various uses, since it does duty for what had been three cases in IndoEuropean, the true ablative, the locative and the instrumental; for vestiges of the locative, see Chapter 15 and, for the instrumental use of the ablative, see Chapter 16. In his Critical Latin Grammar (1772), John Coleridge suggested that the ablative would be better called the quale-quare-quidditive, the “What sort-why-what” case, but even he himself did not actually use this term. (In his Notebooks [August 3rd 1805], his son, the poet Samuel Taylor Coleridge wrote: “I hold it scarcely possible to be too jealous of the purity of Language, even in technical terms”. Not surprisingly, it does not appear in the OED, though John Coleridge’s great-grandson, Herbert Coleridge, was appointed to be the first editor.) In the classical period, the Romans did not even have a name for the ablative, simply calling it the “sixth” case, or the “Latin” case (reflecting the fact that the ablative does not exist in Greek [Roman grammarians derived most of their terminology from Greek models]; modern Greek textbooks of Latin grammar use a calque on ablativus); the ablative continues to be taught after the vocative in Latin textbooks in e.g. Italy, France and Spain. Julius Caesar is commonly credited with the coinage ablativus, but, although he took a great interest in Latin grammar, this attribution is probably erroneous. It is natural that Latin grammatical terms should be so heavily influenced by Greek; the Romans thought, wrongly, that Latin is derived from Greek.

• Vocative is derived from the verb vocare, “to call”.

Information about nouns is usually given in Classical Latin, as in most dictionaries, in the abbreviated form casa, -ae fem. 1 “house”. From this, we learn the nominative singular, the genitive singular, the gender, the declension and the meaning.

Notice particularly the genitive singular. Since the stem remains the same in all the cases in most declensions, it is possible to infer the full inflection of a noun even if one knows only the nominative singular. In the third declension (see Chapter 8), however, the nominative (and vocative) singular of masculine and feminine nouns and the nominative (and vocative) and accusative singular of neuter nouns are very often formed on a different stem from that of all the other cases. Therefore, in order to decline the noun, it is necessary to know not only the nominative singular, but also the genitive singular, since it is the latter which displays the basic stem. For example, the third declension noun meaning “human being” is homo in the nominative (and vocative) singular, but homin- is the stem to which the endings are suffixed in the nominative and vocative plural, and in all other cases, both singular and plural; we must therefore learn the noun as homo (nom. sing.), hominis (gen. sing.).

It is conventional to provide such information for all nouns, even if, as with first declension nouns, which have the same stem in all their cases, both singular and plural, it serves no essential function. The Romans themselves employed the genitive singular in referring to a noun’s basic stem. It is for this reason that the genitive still comes immediately after the often unpredictable nominative in declension-tables in most modern Latin grammar-books.

The order nominative, vocative, accusative, genitive, dative, ablative was firmly and influentially established in the United Kingdom in 1866 by Kennedy’s Latin Primer, and it remains standard in Latin grammars published there. That order is less efficient, particularly for neuter nouns; since they have the same form in the nominative, vocative and accusative, we must therefore wait for the change of stem, if there is to be one, till the fourth case is given.

We tend to think of the nominative as the most important case – as one might say, for example, “casa is the Latin word for ‘house’”. This is only natural, especially since the nominative case shares the derivation of its name with that of nouns in general (nomen), and it is the casus rectus, the “upright case”, away from which the casus obliqui, the “slanting cases”, decline. Nevertheless, it is the genitive which is the key to understanding inflection. It is worth noting also that nouns in the Romance languages are formed predominantly on the Latin accusative case, not on the nominative.

Almost all first declension nouns decline exactly like puella. Proper names borrowed from Greek are the only significant exception.

It is interesting, if inconsequential, to note the variations in the paradigm for the first declension adopted in different countries. Whereas puella “girl” is generally favored in the United States, mensa “table” is more frequent in the United Kingdom, agricola “farmer” in Germany, and rosa “rose” in Italy, France and Spain.

A joke:

“I know a man with a wooden leg called Smith”.

“What”s his other leg called?”

The humorous point, such as it is, depends on an ambiguity in word-order. The first sentence might be translated into Latin as virum cognovi, ligneo praeditum crure, nomine Fabrum, which means literally “I know a man, endowed with a wooden leg, by name Smith”. Since, however, the words for “man”, “endowed” and “Smith” are all in the same case, and all therefore relate clearly to the same person regardless of word-order, a Roman would not find the joke even remotely funny.

Perhaps very surprisingly, especially given that “the” is twice as common as any other word in written English, with “a[n]” also very prominent, Latin has no definite or indefinite article. In translation, we must supply either “the” or “a” or neither as seems appropriate from the context. The sentence nauta puellam videt can be translated equally well as either “The sailor sees the girl” or “A sailor sees a girl” or “The sailor sees a girl” or “A sailor sees the girl”.

Since Latin has no definite or indefinite article, it lacks the option of zero article, as in “I never eat fish”, as well as “I am eating a fish” and “I am eating the fish”. A language with a gendered article (such as the French singular forms le, la, un, une) can occasionally be at a disadvantage. In detective stories, a marked gender often spoils the challenge by eliminating all the possible male, or all the possible female, suspects. In the Greek version of Agatha Christie’s Why Didn’t They Ask Evans?, Γίατι δέν ῤώτησαν τόν Ἔβανς; (Yati den rotesan ton Evans?), the masculine article τόν (ton) is positively misleading and unfair to the armchair detective.

It is estimated that three-quarters of all languages now spoken also do without articles. All those in the Slavic family except Bulgarian are such languages. Classical Greek, Arabic and Welsh have definite articles, but not indefinite. Interlingua, one of the best attempts to manufacture a universal language (a fashion now mercifully out of favor), is basically Latin without inflection; it has an uninflecting definite article, le, and an uninflecting indefinite article, un.

Educated Romans of the classical period were bilingual in Latin and Greek, and must have been impressed, even intimidated, that an Athenian could say “the” in nineteen different ways, a fact which allows for a rather striking, but perfectly normal, word-order of the type “The farmer hears the (accusative) of the (genitive) in the (dative) sty pig gruntings”. The Roman rhetorician Quintilian notoriously, and rather defiantly, said noster sermo articulos non desiderat “Our language does not need articles”, but Priscian, whose Latin grammar was written for Greeks, calls their absence a deficiency, and the Romance languages developed them from the Latin pronouns ille “that” or ipse “himself” (see Chapter 17) and the adjective unus “one” (see Chapter 10).

Not all Greeks were as critical as Priscian of Latin’s lack of articles:

“The language of the Romans has no articles at all, and uses nouns without decorative fringes (so to speak). This is not surprising, given that even Homer, who is such an expert in the deployment of words, attaches articles to only a few of his nouns, like handles on drinking cups which do not really need them, or crests on helmets” (Plutarch Platonic Questions 10.3).

The definite article in classical Greek and in English likewise developed from pronouns; “the” is a weakened form of “that”. A[n] is derived from the Old English number án “one”. (When the form was weakened to a before a consonant and sometimes written as a prefix to the noun, this led to the false division of words; for example, “an adder” and “an apron” should properly be “a nadder” and “a napron”; conversely, “a newt” should properly be “an ewt”.) Even before Quintilian’s time, grammarians themselves had been using the pronoun hic “this” (see Chapter 17) to facilitate discussion of the various cases and genders, a convenient device which still flourished in Shakespeare’s time: “Articles are borrowed of the pronoun, and be thus declined, Singulariter, nominativo, hic, haec, hoc” (Merry Wives of Windsor IV.1).

Just as we must supply articles as the context demands when translating into English, so also with personal pronominal adjectives. Although Latin has a full range of words meaning “my”, “your”, “his” etc. (see Chapters 6 and 17), they are not used as much as in English. This omission occurs particularly with reference to body parts or family members; for example, agricola fīliam amat “The farmer loves his daughter”, agricola, fīliam amāre dēbēs “Farmer, you ought to love your daughter”. That the omission is easy and natural may be demonstrated by a sentence such as “The woman is holding [her] baby in [her] arms”; she cannot be holding the baby in someone else’s arms, and we readily assume that the baby is hers unless we are told otherwise.

A noun in the genitive usually comes immediately before or after the noun which it modifies. Without this restriction, the possible variations in word-order in sentences such as nauta agricolae filiam videt would be even greater. Prepositions, as their name suggests (praepositum “a thing placed in front”), normally come immediately before the word which they govern. Since, however, there are very few rigid rules, it is hard to go wrong, and beginners should not worry unduly about word-order.

A frequent error among English-speaking students is to assume too readily that the first noun in a Latin sentence is the subject, as it almost invariably is in English. It is true that classical Latin is a S(ubject) O(bject) V(erb) language, but the great flexibility in word-order made possible by the heavy use of inflection means that little reliance can be placed on the general tendency for the subject to come at or near the beginning of its clause, and of the verb to come at or near the end.

It is reasonable to suppose that, in spoken Latin of the classical period, which was in advance of the literary language in its use of prepositions rather than inflection to indicate the function of nouns and pronouns, word-order would have been considerably more comparable to that of the modern Romance languages.

Medieval grammars often tried to help students by constructing Latin sentences according to the word-order of their vernacular language. Exercises in this course, by contrast, in an effort to get away from English word-order, will emphasize very strongly the tendency for verbs to be placed at the end of the clause.

Modern English, like the Romance languages, is SVO, but German, which is very closely related to English, is essentially SOV, even though it frequently uses SVO in main clauses (with stringent rules for word-order quite alien to Latin): contrast Das Nilpferd wird das Schwein umbringen, weil es sein Feind ist “The hippo will the pig kill, because it its enemy is” with Das Nilpferd bringt das Schwein um, weil es sein Feind ist “The hippo kills the pig, because it its enemy is”.

In many languages, the standard order is quite different from modern English; in Old Irish the verb usually comes first, the subject precedes the verb in Beowulf in fewer than 20% of sentences, and SOV is the commonest order in languages in general. The rules for word-order are a primary concern in the teaching of English as a second language: why is “Sometimes I read a book” correct, whereas “Never/Always I read a book” is incorrect?

The potential for ambiguity demonstrated in the various possible interpretations of the form puellae is more a difficulty in theory than in practice. The English word “set”, which can be both a verb, a noun and an adjective, has the longest entry in the OED, with more than 430 senses distinguished for the verb alone; this fact might appall a beginner learning English as a second language, but context almost always ensures that it is understood correctly (provided that the particular sense is known).

Apposition is derived from the participle appositus, “placed beside”. A noun in apposition always agrees in case, but not necessarily also in gender and number, with the noun to which it is in apposition. For example:

Romulus XII vultures (acc. pl. masc.), omen bonum (acc., but sing. and neut., in apposition to vultures), videt.

Romulus sees 12 vultures, a good omen.

vultures and omen are 3rd declension nouns (see Chapter 8); bonum is a 2nd declension adjective (see Chapter 6).

Only a small number of first declension nouns are masculine. Some, such as agricola, nauta and pirata, refer to men in occupations which the Romans regarded as exclusively male preserves. Often, as with nauta, pirata and poeta, they are closely related to or directly derived from Greek first declension masculine nouns (a much larger category than in Latin). There is no Latin word for a female farmer, sailor or pirate. Sappho’s poetry, of which only pathetic remnants survive, was regarded in antiquity as almost a match for that of Homer, but no first-rate Latin poetry was written by a woman, and the words for “poetess”, poetria and poetris [3rd decl.], are both very rare.

To put the rote-learning of Latin nouns in its context, it must have seemed entirely natural in an educational system which still, to some extent, applied the same principles to English grammar. Thomas Dilworth’s A New Guide to the English Tongue (1751), for example, is typical of its time in expecting schoolboys to learn such tables as:

It is frequently said, particularly by Latin teachers, that learning Latin is the best way to improve one’s understanding of English grammar. This may be true, in so far as the two languages function in comparable ways, or in so far as the differences are instructive, but to attempt to explain a language so uncompromisingly uninflected as English in terms of a language so strongly inflected as Latin could hardly be more misleading.

In defence, if not justification, of such grotesque blindness in the teaching of English grammar, it may be observed that the Roman grammarians themselves had sometimes forced the model of Greek grammar on Latin, ignoring basic differences in the nature of the languages. Since Greek and Latin, however, are as similar as Latin and English are different, the consequences were not so grave.

The Latin noun familia has a wider application than does its derivative in the Romance languages and English. It refers to dependents and slaves as well as to relatives by blood, and “household” might be a better translation than “family”. This wider sense is preserved in our word “familiar”, as opposed to “familial”.

In most inflected languages, prepositions are frequently constructed with more than one case. Just as in and sub take the accusative if motion is implied, the ablative if it is not, so there are ten prepositions in German which take either the accusative or the dative according to much the same criteria; e.g. Das Schweinchen kam in das Haus hinein (“The piglet came into the house” [acc.], but Das Schweinchen war im [= in dem] Haus (“The piglet was in the house” [dat.]). As with in and sub, the basic sense of these German prepositions remains the same, regardless of the case of the noun or pronoun with which they are constructed.

In ancient Greek, however, fewer prepositions are limited to a single case than are constructed with two or even three cases (and Greek has only five cases). Moreover, the case of the noun or pronoun can radically change the sense of the preposition; for example, διά (dia) can mean “on account of” with the accusative, “through” with the genitive, πρός (pros) can mean “to” with the accusative, “from” with the genitive, “in addition to” with the dative. Numerous Latin prepositions bear several several distinct or tenuously related senses, but usually a difference of case is not involved; for example, de with the ablative means both “down from” and “concerning”, ob with the accusative means both “against” and “on account of”, pro with the ablative means “in front of”, “instead of” and “in accordance with”, propter with the accusative means both “near” and “on account of”, whereas super with the accusative means “above” or “in addition to”, but with the ablative it usually means “concerning”.

It is important to distinguish the use of ad with the accusative from the use of the simple dative of the indirect object. The former is used when motion is involved or strongly implied (“The farmer throws food to his pigs”), the latter when the emphasis is more on the advantage or disadvantage resulting from the action (“The farmer shows food/gives food/talks to his pigs”). Hence, verbs of giving, saying and showing are constructed with the dative of the indirect object.

The symbol @ is a long established scribal abbreviation for ad. (@ is still without a generally accepted name in English, and the French government’s attempt to impose the abstruse term arrobe has met with little success, though it has affectionate names in other languages: in Czech, it is a zavinac “pickled herring”, in Dutch, an apestaart “monkey”s tail”, in Greek, a papaki “little duck”, in Italian, a chiocciola “small snail”, in Russian, a sobachka “small dog”.) Similarly, & is a scribal abbreviation for et, as is made clear by its name “ampersand” (sc. “and per se – and”), a rather clumsy term, first attested in 1837.