Chapter 12
RULE 1: The perfect tense shows action completed in the past.
RULE 2: The perfect stem is marked by (a) the addition of a w-sound; (b) the addition of an s-sound; (c) reduplication; and/or (d) the lengthening of the base vowel.
I. Grammar
This lesson marks a large step, both conceptually and practically, for students. They will double the number of tenses they know and enter new verb terrain, entailing the fuller application of "principal parts" and different "aspects" of past action. Work carefully and methodically—which does not necessarily mean slowly—through the material: explain, first, the nature of the perfect system; then how the perfect stem is generated; and, finally, how the various perfect tenses are formed and translated. Make sure that students understand the step you have just explained before proceeding.
A. The Nature of the Perfect System
The name perfectum was given to this tense system by ancient grammarians who saw its defining character as "completed action," as opposed to the infectum (the present tense-system including the imperfect). Perfectum means "thoroughly (per-) done (-fectum)"—our word perfect meaning "in a faultless state" is only first attested widely in the fourteenth century CE. To comprehend the Latin verb system fully, students must understand the distinction between Latin im(per)fectum and perfectum.
Consider the following example: "When I was taking a shower [imperfect: continuous/on-going past action], the phone rang [perfect: single-event, stopped past action]." A somewhat similar attitude toward time can be seen in the ablative of (point in) time as distinguished from the accusative of (duration of) time. Just like the ablative of time, all perfect tenses entail in some way "point in time." The perfect also conveys a sense of having impact on the present or being true up to the present time, a connotation the imperfect does not, e.g. "I fell down yesterday (but I'm here and still alive)" as opposed to "I have fallen down just now (and may have broken my leg)." Most students, practical creatures that they are, will want to know only how to generate and translate these tenses, but to appreciate the language fully, they must also understand the essense of the perfect tense system and how it relates to other forms in Latin.
B. The Perfect Base
Wheelock says at the beginning of the chapter that students need not "worry" about conjugations in the perfect system. That is only true of endings. The conjugations have distinctive, "regular" ways of forming their perfect bases, so at least to that extent the distinction between conjugations does apply to the perfect system. True, there is no distinguishing thematic vowel in the perfect system as there is in the present system, nor are any of the perfect tenses formed differently in different conjugations. At the same time, however, most a-stem verbs retain their -a- in the perfect and many i-stem verbs their -i-.
Because the perfect stem is not always predictable from the present stem, it must be memorized for each verb (the third principal part). However, many perfect stems follow a predictable pattern falling into three general categories and a group of "mixed" formation:
- "Regular" perfect forms. Each conjugation exhibits a "regular" way of forming the perfect base. In all but third conjugation, this means adding a w-sound (Latin v) to the end of the present base, cf. Eng. -ed:
lauda- > laudav- (amav-, cogitav-, errav-, servav-, tolerav-, vocav-, etc.)
mone- > monu-, where u represents w-sound (debu-, docu-, habu-, valu-, etc.)
audi- > audiv- (examples from verbs which will be covered later: sciv-, nesciv-)Third conjugation verbs use -s- as a perfect marker so often that as far as students are concerned you can say that -s- is the perfect marker in third conjugation. [For the historical development of s-perfects, see "History of the Perfect" below.]
duc- > dux-, where dux- = *duc/s/- (mis- = *mitt/s/-, scrips- = *scrib/s/-, vix- = *viv/s/-)
- Lengthened vowel. Some verbs, mostly from third, third -io and fourth conjugations, form their perfect base by lengthening the vowel in the stem of the verb:
ag- > êg- (vên-, invên-, cêp-, fêc-, fûg-, vîd-), cf. Eng. come > came, take > took, write > wrote.
- Reduplication. A few verbs form their perfect stem by "reduplicating" the first sound in their base, i.e. "doubling the sound," a sort of syntactic stuttering:
da- > ded- (examples from verbs which will be covered later: tang- > tetig-, curr- > cucurr-, pell- > pepul-, disc- > didic-)
- Mixed Formation. Some verbs use a combination of the forms above to create their perfect stem:
sent- > sêns- (s-marker from third conjugation + lengthened vowel)
and among other verbs students will soon learn:
intelleg- > intellêx- (s-marker from third conjugation + lengthened vowel)
trah- > trâx- (s-marker from third conjugation + lengthened vowel)
iung- > iûnx- (s-marker from third conjugation + lengthened vowel)Some verbs borrow a certain way of forming their perfect bases which is distinctive to a conjugation other than their own:
remane- (II) > remans- (as if from III)
possum (irreg.) > potu- (as if from II)and among other verbs students will soon learn:
iube- (II) > iuss- (as if from III) [but see s-perfects below].
rapi- (III -io) > rapu- (as if from II)
pet- (III) > petiv- (as if from IV)Some irregular verbs derive their perfect bases from other verbs (a process called "composite conjugation") and show a conflation of different bases:
*es- (sum, esse) > fu-
and among other verbs students will eventually learn:
fer- > tul- (the perfect base of tollo)
toll- > sustul- (the perfect base of sub-tollo)
These categories will help students memorize verb forms and, more importantly, later recognize perfect forms. I always call to students' attention that if they see a verb form with reduplication or a w-sound it is likely to be perfect (although there are exceptions: iuvo, vivo, nanciscor, titubo). These are the very things which signalled the perfect tense to the Roman ear, just as when we hear an -ed or a different (usually longer) grade (= vowel) of verb stem in English, it is our signal that the verb is past tense [but note that there are exceptions in English, too: wed, bed, tread, etc.].
C. Formation and Translation of the Perfect Tenses
After learning the perfect base, students can create a full verb form in the perfect tense by adding perfect endings. This process is somewhat simpler than forming a verb in the present-tense system, inasmuch as perfect-tense verbs have only two elements (base + ending) and are less often irregular. Students must memorize the endings for the three perfect tenses and learn the simple process of adding them to the perfect base. A few, clear demonstrations should suffice. Only one set of forms may give them any real problem: the -erunt in the perfect vs. the -erint in the future perfect. If the future perfect endings followed the future forms of sum strictly, the future perfect should have -erunt in the third plural, but because -erunt became the third plural ending in the perfect (for which, see below) the future perfect ending was regularized to -erint along the lines of -eris, -erit, etc.
The translation of perfect-tense forms is also relatively easy. Lacking a simple past (saw vs. have seen, cf. the Greek aorist), Latin uses the perfect tense to encompass various ways of looking at past action ("aspects" of the past): the simple past (saw), the affirmative (did see), and the perfective (have seen, i.e. having special impact on the present).
To explain the relationship between the tenses, I have often found a timeline useful:
[Time flows from left to right.]
| PAST | Present [+0] | FUTURE | ||
|---|---|---|---|---|
| Imperfect [-1] | Future [+1] | |||
| Pluperfect [-2] | Perfect [-1] | Future Perfect [+.5] | ||
Consider Oedipus:
- PRESENT [+0]: Today I am putting out my eyes with my wife's brooches.
- IMPERFECT [-1]: Because I married my own mother.
- PERFECT [-1]: And we have had four children together.
- PLUPERFECT [-2]: But before that I had killed my father.
- FUTURE [+1]: So I will go to Athens and be translated into heaven (according to Sophocles).
- FUTURE PERFECT [+.5]: But before arriving there, I will have wandered around Greece for many years.
The pluperfect and the future perfect are the hardest tenses for students to grasp. Explain that the pluperfect represents "two steps" back in the past—in order to have a pluperfect, there must be a past reference point from which the pluperfect ("more-completed") shows action happening even further back in the past: "I didn't do my homework yesterday and I hadn't done it the day before, either." The future perfect shows action prior to the future: "By the time I will turn my homework in, I will have missed the deadline by two days." I also point out to students that the tenses of the perfect system are ordered in such a way as to imitate those of the present:
| Time Frame | Present System [+0] | Perfect System [-1] |
|---|---|---|
| Ground Zero [+0] | Present [= +0] | Perfect [= -1] |
| One step backward in time [-1] | Imperfect [= -1] | Pluperfect [= -2] |
| One step forward in time [+1] | Future [= +1] | Future Perfect [= +.5] |
This relationship between present and perfect tenses will be important not only now in learning perfect endings (the pluperfect endings resemble the imperfect forms of sum and the future perfect endings resemble the future of sum), but also later in learning other forms: the passive voice where sum/eram/ero will again be used; the subjunctive for which there is no future or future perfect; and sequence of tenses in which the present and perfect subjunctive follow primary tenses and the imperfect and pluperfect follow secondary. Thus, laying a solid foundation now will reap benefits in several lessons to come.
D. The Principal Parts of the Verb
Wheelock introduces the concept of the four "principal parts" of the Latin verb before introducing the concept behind the fourth principal part (the perfect passive participle). You can either explain what the perfect passive participle is—which is what I do because in this way the perfect passive participle is not such a shock when it is introduced later—or tell students just to memorize it and not ask questions about its meaning and use.
I point out to students that each principal part conveys essential information about the Latin verb. The First Principal Part is the "dictionary form," the form under which one looks up the verb in the dictionary or, in this case, the back of the book. The Second Principal Part dictates into which conjugation the verb fits, important information for determining how the future and possibly also the perfect tense is formed. With the removal of the final -i, the Third Principal Part shows the perfect stem from which all (active) perfect forms are created. And finally the Fourth Principal Part represents the perfect passive participle, "having been X-ed". I tell students that they are not responsible for anything more than memorizing the Fourth Principal Part, but if they can understand and translate it now, they will be that much further ahead.
E. The History of the Perfect Tense System
The Indo-European perfectum was formed by reduplication, as Greek does regularly, preserved in Latin only in a few "irregular" verbs (dedi, tetigi, pepuli, etc.). Note that in Latin reduplication is often lost or reduced in compound verbs (reddi, contigi, impuli). The Latin long-vowel perfects originated in various ways: some are reduplications which have lost their reduplicated first letter and lengthened the vowel to retain syllable quantity (fûgi, lîqui); some correspond to other Indo-European languages and therefore may be inherited (vêni, lêgi, êdi); and others derive from strong (second) aorist forms (fêci, cf. Greek ethêka; iêci, cf. Greek hêka). The s-perfects (dixi, duxi, scripsi) correspond to (Greek) sigmatic aorists. Thus, the Latin perfect represents a conflation of the Indo-European aorist and perfect tenses, which corresponds to its double use by the Romans as both the simple and perfective past tense.
Normally, perfects in Latin lose the -n- originally inserted into the base as a distinctive present-tense marker (relinquo vs. reliqui, vinco vs. vici); however, some retained it as part of the process of regularizing the language over time: (iungo vs. iunxi, cf. iugum).
The -vi (and -ui) perfects are distinctive to Latin (cf. the distinctive k-perfects in Greek) and may have originated in forms like fui (*fuvi), where the w-sound was originally part of the base and the verb was misanalyzed as fu-vi. The -vi was later extended to other verbs.
- Perfect Endings. The Indo-European endings for the perfect tense were:
| Person | Singular | Plural |
|---|---|---|
| First Person | -a | -imus |
| Second Person | -tha | -te |
| Third Person | -e | -ro(?) |
First Person Singular. Greek retains the Indo-European active perfect ending -a. The Indo-European middle ending -ai became -i (Old Latin -ei).
Second Person Singular. -thai was a by-form of the Indo-European ending -tha which Greek again represents in some verbs, e.g. oistha. Under the influence of the first person singular ending -i, Latin changed -thai to -ti. The -is- preceding -ti is found also in the second plural, -istis, and the infinitive, -is/se, as well as the pluperfect *-isam > -eram and the future perfect *-iso > -ero. This -is- originated in s-aorists and was extended by analogy to other verbs.
Third Person Singular. In Italic the secondary ending -t was added to -e. -et which evolved into -ed in Old Latin (e.g. feced) and later -it in classical Latin when the primary ending -i- was regularized throughout the perfect (except in the third person plural).
Second Person Plural. = -is- (as in the second person singular) + -tis renders the regular second person plural ending.
Third Person Plural. An amazing mess, there are three forms attested in Latin: -erunt, -êrunt and -êre. -erunt is probably from -is/ont by way of rhotacism, with the same -is- as in the second person singular. -êre represents an r-form found in no fewer than six other Indo-European language groups. Latin -re probably developed from the primary *-ri, cf. mare < *mari. -êrunt is, no doubt, a conflation of -êre and -erunt.
- Pluperfect and Future Perfect Endings
See above (Second Person Singular). The -a in the pluperfect represents the past tense. The -i in the future perfect is the same short-vowel subjunctive which became the regular future in Latin (see Chapter 5, "The Future Tense").
F. Handout
Click here for a worksheet on perfect forms.
II. Vocabulary
- deus: Di and dis are contracted forms. The Indo-European base was *deiwos. Early Latin lost -w- (-v- in Latin script) when it preceded -o, as in deorsum (from devorsum) and secundus (from *sequondos). But later the w-sound was restored if there were other forms of a word in which the w-sound did not precede an -o, e.g. sequontor > secuntur > sequuntur (by analogy to forms like sequitur); equos > ecus > equus (by analogy to equi). Words with only a w-sound + -o in all forms lost their w-sound permanently (secundus, deorsum). *Deiwos went both ways in Latin: it both lost its w-sound (*deiwos > *deios > deus) in the noun and regained it in the adjective (*deiwos > *deios > divus), cf. *sei-ve > *se-ve > seu/sive. The plural forms of the noun, di and dis, are the regular forms derived by the contraction of diphthongs: *deiwoi > *dêê > *dê > dî. The disyllabic form, dei, with restoration of e on the analogy of other forms, occurs in later poets.
- libertas: Students should recognize that this third-declension noun ending in -tas must be feminine in gender (see Wheelock, page 32, note 2).
- rex: This word produces many interesting derivatives (Eng. bishop-ric, Ind. rajah), and some fascinating cognates (Gaul Dumno-rix) The last example (Dumnorix) is part of the proof that Latin in general is more closely tied to Celtic than other western Indo-European languages like Greek. Our word rule is a derivative of the Latin diminutive regulus ("a straight stick") coming into English through Old French (reule). The root of the noun base reg- meant originally "to stretch out, make straight" and derived its connotation of governance from a metaphor resembling that behind pax (Chapter 7). The basic metaphor of justice and judgment (i.e. the proper separation of opposing claims) is that a "straight" line divides fairly a field claimed by rival parties.
- dico: From a base meaning "point out," cf. Gk. deiknumi, Lat. indico (cf. indicative).
- vinco: The -n- (technically, a nasal infix) is a present-tense marker (used throughout the present-tense system) which normally disappears outside of the present system, cf. relinquo vs. reliqui, Gk. lambano (with two nasal infixes: la-n-b-an-o) vs. elabon. Therefore, vici has no nasal infix.
- Asia: Note that in antiquity this name referred to the area we now call Turkey. Because the Romans had little idea of the scope or nature of the huge land mass now called Asia, the name was extended to the entire continent.
- caelum: Of dubious origin, this word is possibly related to cavus and Gk. koilos. The heavens would then have originally been the "hollow" (cf. Gk. chaos, "the yawning").
- Caesar: Point out to students that names in Latin decline and therefore fall into a declensional category. Caesar belongs to, of course, the harder third declension. Few names, however, have plural forms, which are reserved for aristocratic gentes (e.g. the Metelli, the Gracchi), the most special emperors (the "Caesars") and, presumably, the odd schizophrenic. Stress the correct spelling of the name—not Ceasar or Seizure!—for some reason this name is hard for students to spell properly. Nor is this a bad opportunity, if you have time, to discuss the Roman system of nomenclature (praenomen, nomen, cognomen). [Caesar's cognomen was in origin dialectal, therefore the Roman rule of rhotacism (intervocalic -s- > -r-) did not affect it.] For formation of the nominative singular, see Chapter 7.
III. Sentences
Practice and Review
- Be sure students know why in hanc viam cannot mean "in this road," but "into this road."
- Note that there is no long mark on the i- in idem, therefore it must be neuter (nominative/accusative).
- Point out to students that "to him" cannot be represented by the dative case in this sentence.