# When is the genitive used for money rather than the ablative?

In some exercises I am doing, there are the following sentences:

Haec stola octo milibus nongentis nummorum Iaponensium constat.

Caro octigentis nummis Iaponensibus constat.

I don't understand why the units of money should be in the genitive in the first example, but the ablative in the second.

• Does this answer your question? Genitive vs Ablative of Price Sep 19, 2021 at 0:33
• @brianpck No. It is not clear at all that the first sentence is an expression of "value" and the second is of price. Both appear to be of price only, so it is not clear what the difference is that two different cases would be used. Sep 19, 2021 at 0:38
• Yeah, I was paying too much attention to the title of the question. In fact, the issue here is entirely different: as Sebastian points out, both sentences use the ablative of price, and the genitive in the first is something completely different, i.e. "worth thousands (abl.) of coins (gen.)." Sep 20, 2021 at 18:55

I don't know what exercises these are, but I believe that octo milibus nongentis nummorum is incorrect. It should be octo milibus nongentis nummis.

These are the rules to my knowledge:

• Numerals are generally adjectives and thus agree with the noun, which in your example is in the ablativus pretii. Therefore you have octingentis nummis. Most of them are not declined, so it's septuaginta nummis etc., but the point stands.

But mille is special.

• Singular mille can be used as a noun (also not declined). So it stands in the ablative or whatever case the context calls for (which is invisible because of the lack of declension) and the counted objects in the genetive (the genetivus partitivus). Therefore you have mille passuum, mille militum, etc., and of course mille nummorum = a thousand sesterces.

• But singular mille is also often used as an undeclined adjective like other numerals, so you can say mille nummi as well.

• Plural milia is always a noun, requiring the counted objects to be in the genetive, and always declined as well. So it's duobus milibus nummorum constant etc.

• But if milia is followed by a smaller numeral, and then the counted objects, those objects are in the same case, so you say: duo milia nongenti nummi, duobus milibus nongentis nummis, but: nummorum duobus milibus nongentis or: duobus milibus nummorum et nongentis etc.

This last rule was not applied here.

• Still, for the last point, Gildersleeve & Lodge §293 gives the following example from Livy 22.7.3 that provides an exception to the usual rule: duo milia quingenti hostium. So the exercises that are referred to in the main question aren't totally wrong in the end – though I suspect this is just a happy accident. Sep 18, 2021 at 23:06