I think that the ancient Roman authors did not always follow the rules of Latin grammar as the are found in a text book. Below I give a quote from Cornelius Nepos’ biography of Hamilcar. As far as I can see it breaks the well known sequence of tenses rule of Latin grammar.
“Nam diuturnitate externi mali tantum exarsit intestinum bellum ut numquam in par periculo fuerit Carthago, nisi cum deleta est.”
Since “exarsit” is an historical (past) tense, it should be followed by an “ut” clause with either an imperfect subjunctive or a pluperfect subjunctive. But “fuerit” is a subjunctive of the perfect tense.
Well, if there is a fault in my logic, I hope that somebody will point it out for me.
I think there might be a bit of a fault in your logic in terms of thinking that ancient authors conceived of ‘rules of grammar’ in the same manner we do. It seems that rather than strict rules, there were standards or conventions of writing.
Well now, after Googling “Sequence of Tenses”, I see that Bedwere is correct. I was basing what I said on textbooks that present the rule of Sequence of Tenses without including the less common usage of the perfect subjunctive in place of the imperfect subjunctive. But on the internet are one or more sources of Latin grammar that include the use of the perfect subjunctive, used as described above. Well, I was wrong. I guess I will have to keep looking if I want to try to find any grammatical usage by classical authors which does not fit with the rules of Latin grammar.
Actually, I wasn’t trying to imply anything about the existence or non-existence of rules of grammar in the days of ancient Rome. I was just comparing the actual usage of a classical author with rules of Latin grammar as they exist now-a-days. But did the ancient Romans analyze their language and develop rules of grammar? It seems likely enough, but I’ve never read anything about that.
Modern Latin grammars are formulated from the writings of ancient authors, particularly Caesar, Cicero, and Vergil. Here is a nice intro article on ancient grammarians:
The way in which Latin was used (what we might call Latin Grammar) certainly changed over time from say Plautus to Tacitus to choose a period which you might loosely call “classical”. (“Classical” is a very moveable feast).
Most modern grammars as Barry has said are based on the latin authors and some place great weight on Cicero because of the sheer volume of his surviving work. A grammar is not a rule book about how a language must be written, it is a description of how a language has been written.
If you are seriously interested in this I recommend “A New Latin Syntax”, E. C. Woodcock, Bristol 1959 which tries to take account of the historical order of syntactical development in its presentation of Latin Syntax. You might also find The “Blackwell History of the Latin Language”, James Clackson and Geoffrey Horrocks, Oxford 2007 useful. The book Barry links to looks interesting.