[...] Leonardo da Vinci became renowned for his multiple talents: he was a painter, architect, engineer, mathematician and inventor.

In the above sentence, why has only one "a" been used, even for words that begin with a vowel?

  • Related: ell.stackexchange.com/q/13307/3281. Commented Jun 3, 2014 at 21:00
  • Thanks for the link. However, I'd like to know if there is any specific rule to explain the reason why this happens.
    – M.N
    Commented Jun 3, 2014 at 22:48
  • 3
    This is called "Conjunction Reduction", the syntactic rule that deletes repeated material in conjoined clauses. I tried to find its formal explanation, but I couldn't find it. J.R.'s comment in the link I mentioned above is good. You might also find John Lawler's answer useful too: english.stackexchange.com/questions/114646/…. Commented Jun 3, 2014 at 22:59
  • See also: english.stackexchange.com/questions/51930/…. Commented Jun 3, 2014 at 23:04
  • 4
    Essentially, the reason that this works is that conjunctions can coordinate items at any level, as long as they are the same type... in this case, the bare nouns are being coordinated, and all of them are dominated by the same determiner phrase, and sister to the same determiner.
    – jimsug
    Commented Jun 4, 2014 at 0:55

2 Answers 2


Leonardo da Vinci became renowned for his multiple talents: he was a painter, architect, engineer, mathematician and inventor.

Your example works because:

  • you can use and then:
    • In speech, elide all conjunctions, up to (and arguably, including) the last;
    • In writing, replace all but the last conjunction with commas
  • When you join units at the Noun level, the Determiner only needs to "agree" with the first in the series, as it's arguably for phonotactic reasons - what sequences of sounds are permissible in English, which the writing system has come to reflect.
    The more abstract, implicit "Determiner" can simply have the same meaning of the explicit Determiner, the:

    • Singular
    • Indefinite (a/an rather than the)

    Not that it is either a or an, but just the above information

  • and because conjunctions can coordinate - join - units at any level - and the result is the same unit:
    • Noun and/or Noun forms another "Noun"
    • Noun Phrase and/or Noun Phrase forms another "Noun Phrase"

In your example, what is happening is something like (let's ignore the final comma, for the purposes of this demonstration:

Det  = Determiner
Cnj  = Conjunction
Noun = Noun
[]   = Elided/replaced Conjunction

Leonardo da Vinci... a   painter and architect and engineer and mathematician and inventor.
                     Det Noun    Cnj Noun      Cnj Noun     Cnj Noun          Cnj Noun


Leonardo da Vinci... a   painter, architect, engineer, mathematician and inventor.
                     Det Noun   []Noun     []Noun    []Noun          Cnj Noun

Another possible valid conjunction would be, as you point out one that has Determiners for each Noun (forming a Determiner Phrase (more commonly known as a Noun Phrase), or Nominal Group):

NP = Neterminer Phrase
DP -> Det + Noun

Leonardo da Vinci... a   painter, an  architect, an  engineer, a   mathematician and 
                     Det Noun     Det Noun       Det Noun      Det Noun          
                     NP         [] NP          [] NP         [] NP               Cnj
an  inventor.
Det Noun

You might need to be a bit imaginative with the above interlinear explanation. The plain text is the first line, and the second line, as with the above examples, is the word class of each constituent - I'm analysing the commas as realising a grammatical function here. The third line is a higher-level of categorisation.

Now, if you're technically inclined, my understanding is that generative grammar doesn't allow for a node to have two parents, and so you couldn't have paratactic (equal) relationships between each of the above noun phrases or nouns. You would be forced to have a Noun Phrase formed by two coordinated Noun Phrases, which are themselves formed of two coordinated Noun Phrases... if you have Noun (Phrases) which are not a power of 2 in number, then you're forced to place one above another. I'll gladly edit if I've misunderstood.

  • 3
    *cries* Until this day, innocent ELL had not been exposed to the DP hypothesis. Oh, for happier days now gone!
    – user230
    Commented Jun 6, 2014 at 21:00
  • I suppose I should point out that DPs, as well as NPs, nominal groups, etc, are all just hypothetical - there is literally no way to prove that they exist as entities, beyond their usage as a mediating language in explaining the realisations of lexicogrammar in discourse.
    – jimsug
    Commented Jun 6, 2014 at 23:55

A simpler example is: 'I eat with [a knife] and [a fork]' > 'I eat with a [knife and fork]' (which emphasises that I use the knife and fork together) compared to 'I eat with [a bowl] and [a spoon] (which emphasises that those are two different kinds of utensils' > ?'I eat with a [bowl and spoon]'.
Saying 'he was [a painter], [an architect], [an engineer], [a mathematician] and [an inventor]' might emphasise that he was all of those things separately, but part of his genius was that he was often mixtures of those things, or even all of those things (and more) at the same time.

  • Are you saying "I eat with a bowl and a spoon" is different from "I eat with a bowl and spoon"?
    – M.N
    Commented Jun 7, 2014 at 10:41
  • I think there is no real difference between 'a bowl and a spoon' and 'a bowl and spoon'. Including 'a' or 'an' each time, or leaving it out each time, is a matter of choice - some people do, other people don't. If they leave it out, it is most likely to be because the original sentence was very long (like the Leonardo sentence) or because the ideas are closer together (like 'a [knife and fork]'). If they include it, it is most likely to be because the ideas are further apart (like '[a bowl] and [a spoon]').
    – Sydney
    Commented Jun 8, 2014 at 2:25

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .