Why is the past tense in Russian so odd?

The past tense of verbs in Russian looks very odd. It marks the gender and number of the verb’s subject, but does not mark whether the subject is 1^st person (I / we), 2^nd person (you) or 3^rd person (she / he / it / they). In this respect, the Russian past tense differs from other Russian tenses and from tenses in other languages of the Indo-European family.

This post examines how the Russian past tense came to be like this, and how its ancestor is now reflected in other Slavonic languages. The post also looks at the counterparts in the modern Slavonic languages of the ancestor of the copular verb byt’.  

Russian inflection

Russia is an Indo-European language. In many Indo-European languages, the form of verbs changes to show the number and person of the verb’s subject. This change in the form of a verb is called inflection.

Inflection in present tense

In the present tense, Russian verbs do indeed inflect for both number and subject. As an example, table 1 shows the present tense of the verb delat’, meaning do.

1. For convenience, I give all examples in the Roman alphabet, not in Cyrillic.
2. I use a hyphen to separate the verb’s present tense stem (delaj) from the inflectional ending.
3. <š>stands for the last consonant in English mesh, represented by <sh> in English writing and by the symbol ʃ in the International Phonetic Alphabet.

Inflection in past tense: modern Russian

But in the past tense it is different. Russian verbs inflect only for gender and number. And in the plural, they don’t even inflect for gender, as table 2 shows.

Inflection in past tense: old Russian

Why does the past tense inflect like this in modern Russian? The reason is that the form now serving as the (only) past tense in modern Russian was a past participle in old Russian. That participle combined with the present tense of an auxiliary verb—the copular verb byt’ (‘be’)—to form the perfect tense. Like all other verbs, the present tense of byt’ inflected for person and number. And like adjectives, the past participle inflected for gender and number (singular / dual / plural: modern Russian has lost the dual.)

As an example, in old Russian, the form esm’ kupila (‘I (have) bought’, spoken by a female) consists of esm’ (1^st person singular present of byt’) and kupila (feminine singular form of past participle of kupiti, ‘buy’).

Lost present tense of the verb byt’

Russian gradually lost the present tense of the verb byt’, both when used as a copula and when used as an auxiliary. Thus, ‘[I] bought’ (spoken by a female) is simply ‘ [ja] kupila’ in modern Russian. Vlasto (1986) says that the 3^rd person auxiliary had already become only optional by the 12^th century.

The last vestige of byt’ in the present tense

In modern Russian, the only remaining form of the copula is the 3^rd person singular est’ (and occasionally in scientific texts, the old 3^rd person plural form sut’). This is now used only in very limited contexts in some existential or possessive constructions or in tautological definitions.

Cubberley (2002) notes that no other Slavonic language uses est’ in possessive constructions. He comments that this construction may have come into Russian through contact with speakers of Finnic languages. A similar construction in Finnish uses the adessive case How many cases are there in Hungarian and Finnish? (2) – Language Miscellany  

Other Slavonic languages

There are interesting variations in how the old Slavonic perfect form has evolved in the other modern Slavonic languages. I discuss below:

• form of the l-participle
• the copula
• other past tenses
• other points

Form of the l-participle

Earlier forms of the Slavonic languages formed the perfect tense using a participle—often called the l-participle because it contained the consonant [l]. All the modern Slavonic languages still use the l-participle in a past tense.

Singular of the l-participle

All the modern Slavonic languages produce the singular forms of the l-participle with the suffixes -l (masculine), -la (feminine), -lo (neuter).

In some languages, the -l suffix of the masculine singular has become a vowel (-o in Bosnian/Croatian / Montenegrin / Serbian; -u in Belarusian) or vowel-like (-v in Ukrainian).

Plural of the l-participle

In the plural, Russian and Bulgarian have a single ending (-li) for all forms of the l-participle. Ukrainian has -ly and Macedonian has -le.

Bosnian/Croatian / Montenegrin / Serbian has retained separate plural forms for the 3 genders: -li (masculine), -le (feminine), -la (neuter). Old Russian also had distinct gender endings in the plural: -li (masculine); -ly (feminine); -la (plural), but these have merged in modern Russian.

Polish and Czech make a slightly different distinction in the plural, based on what is sometimes called secondary gender. Although both those languages draw a line, they draw the line in slightly different places:

• Czech still distinguishes 3 genders in the plural: -li (for masculine animate nouns); -ly for inanimate masculine and feminine (animate or inanimate) nouns; and -la for neuter nouns.
• Polish distinguishes 2 genders in the plural: -li (for nouns denoting masculine humans—the ‘virile’ sub-gender); and -ły for all other nouns.  

In Polish (and Sorbian), the ending -li is written with <l> when this consonant is palatalised, but is written <ł> in the endings -ł, -ła, -ło and -ły to show that it is unpalatalized (hard) there.

Copula

A copula verb (such as ‘be’) links the subject of a sentence to a noun or adjective. As noted above, Russian has largely lost the present tense of the verb byt, both in its use as a copula and as an auxiliary to form the perfect tense.

Like Russian, the other East Slavonic languages occasionally use the copula in an invariant form: est’ (Russian); je (Ukrainian); ësc’ (Belarusian).

The other Slavonic languages have all kept the present tense of their equivalent of byt’, both as a copula and as an auxiliary, though to varying degrees. Table 3 shows the present tense of byt’ in 5 languages:

Notes on the forms of the copula and auxiliary:

• The forms shown for Bosnian/Croatian / Montenegrin / Serbian (BCMS) are the full forms used when the copula is stressed. The following reduced (clitic) forms are used (1) when the copula is unstressed and (2) as the auxiliary: sam / si / je / smo / ste / su. Except for 3^rd person singular je, the reduced forms are the full forms without the initial je syllable.
• The Bulgarian forms are similar to the BCMS clitic forms.
• Slovene has forms similar to Slovak, but sem, smo, so instead of som, sme, su. Slovene also retains a dual, in addition to singular and plural.
• In Czech, the forms written with <js-> are generally pronounced with [s-].
• The above forms are all noticeably similar to the forms used in Old Russian and believed to have existed in Late Proto-Slavonic (esm’; esi; est’; esm; este; sut’)
• A number of the languages have negative forms of the copula, generally starting with [n-]. For example: Russian (net, negative of est’)

Notes on use of the auxiliary

• As noted above, Russian now forms the past tense only with the l-participle, with no auxiliary. The same is true of the other East Slavonic languages (Ukrainian and Belarusian).
• Polish, Czech and Slovak generally use the 3^rd person forms (singular and plural) as copulas but not as auxiliaries. Thus, for the 3rd person, the past tense in those languages consists of an l-participle without an auxiliary.  
• Polish may omit the copula in some present tense contexts, replacing it with to or with a dash. (In some case, Russian too uses the dash, sometimes followed by the demonstrative eto).
• Polish has reduced the auxiliary to an affix, which often attaches directly to the end of the l-participle, but sometimes to an earlier word in the sentence.
• In other West and South Slavonic languages, the copula and auxiliary are fully alive.
• In many of the Slavonic languages, the auxiliary is a clitic. This means it cannot stand by itself as a fully independent word, and must instead appear next to a word that is fully independent. I will not attempt to summarise where the clitic may appear, as the requirements are complex and differ from language to language.  

Past tense personal affixes in Polish

The Polish affix is derived from the endings on być (‘be’):  1^st person (-m, -śmy) and 2^nd person (-ś, -ście). There is no affix for the 3^rd person.

An example containing one of those endings is: śpiewałem (‘I sang’, said by a male person). The structure of this form is verb (śpiewa) + participle marker (-ł) + vowel marking gender (-e) + personal ending (-m). Similarly, you (feminine singular) sang is śpiewałaś (śpiewa + ł + a + ś).

In contrast, in the Czech equivalents of those forms, the auxiliary appears as a separate word after the participle: zpíval jsem; zpívala jsi

In some cases, instead of attaching to the l-participle, the Polish affix attaches (as a clitic) to the end of an earlier word in the sentence, as in the following Polish example:

(The hyphen is used here to show where the affix attaches. Normal spelling does not include this hyphen.)

Another possible order for this sentence is the following, with the affix attaching to the first word:
Czarnego-ś psa widział?

If the clitic is attached to the verb, this sentence is:
Czarnego psa widział-eś?

Other past tenses

Alongside the perfect tense using the -l participle, Proto-Slavic had several other past tenses. They included the imperfect (for repeated, habitual or ongoing past actions), the aorist (for single or completed past actions) and the pluperfect.

The East Slavonic languages have retained only the simple past tense descended from the old perfect.

Bulgarian has retained several past tenses and they now form a complex system. Bosnian/Croatian / Montenegrin / Serbian kept some other past tenses, though they are falling out of use.

Polish, Czech and Slovak and Slovene have lost the imperfect and aorist, and the pluperfect is archaic in Polish.  

Other points

Some Slavonic languages combine the -l participle:

• with the future tense of the copula to form a pluperfect tense;
• with the future tense of the copula to form a future perfect or other future tense;
• with an inflected form of the copula (or with an invariable particle derived from the copula) to form a conditional form.  

The ancestral languages Indo-European and Proto-Slavic distinguished 3 numbers: singular, dual and plural. Of the modern Slavonic languages, only Slovene and Sorbian (spoken by around 30,000 people in eastern Germany) have retained a dual.

Sources

The Slavic Languages, by Roland Sussex and Paul Cubberley (2006)

The Slavonic Languages, edited by Bernard Comrie and Greville G Corbett (1993)

Russian: a Linguistic Introduction, by Paul Cubberley (2000)

A Linguistic History of Russia to the End of the Eighteenth Century, by AP Vlasto (1986)

Slavische Interkomprehension: Eine Einführung, by Karin Tafel, Rašid Durić, Radka Lemmen, Anna Olshevska, Agata Przyborowska-Stolz (2009)

A Handbook of Slavic Clitics, by Steven Franks and Tracey Holloway King (2000)