In both Latin and Greek, the most common citation form for a verb is the first person singular present indicative active. In other words, dictionaries will generally be indexed by amō and λύω rather than amāre and λύειν. There are some exceptions, but this seems to be the most common practice.

Why is this? Is it a modern practice, or did the ancients cite words by their 1S too?

(I know why the 1S (first person singular) form is included as one of the principal parts in Latin, because it distinguishes the third -ō verbs from the third -iō verbs. But that doesn't explain why it's the citation form—after all, there are three other principal parts too, none of which are used as the citation form. And in Greek, I can't think of any information the 1S provides that the infinitive doesn't.)

  • 1
    The accepted answer to this question explains why the 1S is one of the principal parts, but not why it's the citation form. The comments have an example from the fourth century CE, which shows that this convention is pretty ancient, but I'm interested to see what other evidence there is.
    – Draconis
    Commented Mar 5, 2021 at 15:41
  • Perhaps there's an answer already there latin.stackexchange.com/a/2005/39
    – Alex B.
    Commented Mar 5, 2021 at 16:41
  • @AlexB. Very true, but that one mentions the Romans also used third singular and infinitive. It seems like it would also be possible that, for example, some prominent post-classical lexicographer made the decision when curating a Latin dictionary (based on Varro or whatever) and that set the standard for many later works.
    – Draconis
    Commented Mar 5, 2021 at 16:58
  • 3
    Your question is very interesting but I'm afraid it would take weeks of research to properly answer it. If/when I have some free time, I might dig deeper into this, but I can't promise anything more definite. Off the top of my head, I'd say it's the first person singular because it's the first form students had to memorize, amo-amas-amat etc.
    – Alex B.
    Commented Mar 5, 2021 at 17:03
  • 2
    Meanwhile, you may want to peruse Donatus' Ars Grammatica (esp. pp. 359-360) archive.org/details/grammaticilatini04keil/page/358/mode/2up
    – Alex B.
    Commented Mar 5, 2021 at 17:23

4 Answers 4


For practical reasons, I imagine the index form is often (but not always) given in the first-person since it comes first in the principal parts and is usually the first in a paradigm memorized.

There is however ancient precedence. In discussing Latin verbs, Varro chiefly (but again not always) gives the first person unless he was specifically discussing another person. See e.g. De Lingua Latina 9.96-100, 105, 107, etc.

Ultimately I think it's just familiarity and convention.


Let me try to approach this from a slightly different angle: What would work as a citation form?

A good citation form would be such that you could deduce all other forms (from the present stem) from it. I will look into the six personal forms of present active indicative and the present active infinitive. The exercise can be extended to other forms, but these are a natural set of candidates and the conclusion will remain the same.

For reference, here are these forms of the five regular conjugation patterns:

  I II III-o III-io IV
inf. laudāre vidēre trahere facere audīre
1S laudō videō trahō faciō audiō
2S laudās vidēs trahis facis audīs
3S laudat videt trahit facit audit
1P laudāmus vidēmus trahimus facimus audīmus
2P laudātis vidētis trahitis facitis audītis
3P laudant vident trahunt faciunt audiunt

Each has its problems:

  • If you are only given the infinitive, you cannot tell the difference between III-o and III-io. [Nor between II and III if there are no macrons.]
  • If you are only given the first person singular (1S), you cannot tell the difference between I and III-o nor between III-io and IV. (You might also confuse III-io or IV for I or III-o, as in facio > *faciare or audio > *audiere. This is ruled out if you know that certain stems are impossible, but this is a potential source of confusion for students.)
  • If you are only given the 2S, you cannot tell the difference between III-o and III-io. [Nor between III and IV if there are no macrons.]
  • If you are only given the 3S, you cannot tell the difference between III-o, III-io, and IV.
  • If you are only given the 1P, you cannot tell the difference between III-o and III-io. [Nor between III and IV if there are no macrons.]
  • If you are only given the 2P, you cannot tell the difference between III-o and III-io. [Nor between III and IV if there are no macrons.]
  • If you are only given the 3P, you cannot tell the difference between III-io and IV.

The conclusion is that no single form will work perfectly. You can, however, dissolve all ambiguity if you are given two forms or one form and the conjugation number. For example, you could give the pair (2P&3P) (eg. trahitis & trahunt and facitis & faciunt and audītis & audiunt), and you could find all else from those.

The canonical choices seem to be infinitive with 1S and 1S with conjugation number. These work but are but are not the only ones that do. If the dictionary has no macrons, the chances of confusion increase somewhat, but there are still several pairs of forms that resolve all ambiguities.

If you want to choose a single citation form, it makes sense to pick one with minimal overlap between the different conjugations. But from a practical point of view, it is wise to choose a relatively short form for which cannot cause confusion between two existing verbs.

Conclusion: The citation form wouldn't have to be what it is. Therefore any explanation must have a cultural aspect to it; it is a matter of convention rather than necessity.

Related older posts you might find interesting in connection to this:


This isn't a full answer, but brianpck provides some interesting evidence for how the Romans thought of verbs, in his answer to a related question.

From Varro's De Lingua Latina VI.5.37:

Primigenia dicuntur verba ut lego, scribo, sto, sedeo et cetera, quae non sunt ab ali<o> quo verbo, sed suas habent radices. Contra verba declinata sunt, quae ab ali<o> quo oriuntur, ut ab lego legis, legit, legam et sic indidem hinc permulta.

We call verbs like 'legō', 'scribō', 'stō', 'sedeō', etc "primitive": the ones that don't come from another verb, but have their own roots. On the other hand, "declined" verbs are the ones that arise from another, like from 'legō' [arise] 'legis', 'legit', 'legam'; in this way a whole lot [of verbs] come from the same place.

In other words, the grammarians seem to have thought of a verb paradigm somewhat like a noun paradigm, and cited it by the first form on the list.

This doesn't explain why the 1S has now become so thoroughly predominant in modern times, when (as we can see in brianpck's question) Quintilian also cited verbs by the 3S, but it might help someone else in tracking down the origins of the modern convention.


It's interesting to note that for many ancient Greek grammarians, the infinitive was not a real verb (ῥημα) "on the grounds that it is not part of a basic sentence or λόγος". See 'Whence and Whither Greek Verbal Lexicography and Pedagogy' (at http://bagl.org/files/volume3/BAGL_3-3_Kraeger.pdf)

About the disadvantages of using the 1st singular as citation form: in addition to the ones mentioned by Joonas, there's confusion between 'video' (vidēre) and 'creo' (creāre), and someone might add possible confusion between verbs on the one hand and nouns + adverbs ending on -o on the other.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.