I recently came across the aphorism Nam illa tumultū gaudēns nōn est industria, from Seneca's epistle "On True and False Friendship". As far as I can tell, a literal translation would be "For rejoicing in tumult is not industry"; the standard idiomatic translation is "For a love of bustle is not industry." But I'm confused about what illa is doing in the sentence. Does anyone know why it's there? I can't figure out what it refers to, nor why it's feminine.
I would translate it thus:
For that which rejoices in tumult is not industry.
What may be confusing is the feminine gender of illa: in English (and some other Germanic languages), we'd use a neuter pronoun (that) in constructions such as this where a (to us) indeterminate pronoun refers forward. But, in Latin, the pronoun referring forward to industria agrees with that word, at least in this case. I suspect that illud would also be possible, though I think I'd prefer illa.
Explained in a more general way: when the subject is a pronoun (illa), and the subject complement is a noun (industria), I believe the subject would normally agree with the subject complement, even if the complement comes some ways after the subject, as here. In Germanic languages, I believe such a pronoun-subject is less likely to agree forward with a complement that it hasn't 'seen' yet, and we would be more likely to use an indeterminate (neuter) pronoun for the subject instead (that).
The text continues like this:
Nam illa tumultu gaudens non est industria sed exagitatae mentis concursatio, et haec non est quies quae motum omnem molestiam iudicat, sed dissolutio et languor.
"... and that which judges all movement to be a nuisance is not quiet."
Here the pronoun is haec, agreeing with quies. The relative clause quae... agrees in turn with haec (whence also with quies).
It refers back to the basic idea of being always restless, as expressed by eos qui semper inquieti sunt in the previous sentence. So the basic clause in English is 'That isn't industry.' Although we might expect the pronoun to be neuter, illud, in Latin, a demonstrative is very often assimilated to the gender (and number) of any predicate noun.
The rule as given in Gildersleeve & Lodge, Latin grammar (§211, remark 5) is:
The demonstrative pronoun is commonly attracted into the gender of the predicate...
The following examples are cited:
at negat Epicurus – hoc enim vestrum lumen est – quemquam, qui honeste non vivat, iucunde posse vivere. (Cicero, De finibus 2.22.70)
'But Epicurus – for he is your light - denies that anyone who doesn't live honorably can live pleasantly.'
Here, although the hoc refers back to the masculine singular noun Epicurus, so it seems as though it should really be masculine too (as in the English translation), it's neuter in Latin through assimilation or attraction to the neuter singular predicate noun lumen.
nam quod Cleomedon modo tamquam mediam et tutissimam uobis viam consilii, ut quiesceretis abstineretisque armis, ostendebat, ea non media sed nulla via est. (Livy, Ab urbe condita 33.21.33)
'As for what Cleomedon was just now pointing out as the middle and safest course of a plan for you – namely, that you sit tight and refrain from arms – that isn't a middle course but no course at all.
Here, ea refers back to quod (which encapsulates the whole substance of Cleomedon's suggestion) but is feminine singular through attraction to via.
The same thing happens with haec in the continuation of this sentence: et haec [referring to eos qui semper quiescunt in the previous sentence] non est quies quae motum omnem molestiam iudicat, sed dissolutio et languor.
Relative pronouns often work the same way.