William Gardner Hale, The Art of Reading Latin: How To Teach It

Verbal constructions should be dealt with in a similar way. The possibilities after conjunctions should, in particular, be entirely familiar. Given a quamquam or a quamvis, the student should be able to tell instantly what is coming. Given an antequam, he should know precisely what the two ideas are, either one of which may possibly be in the speaker's mind, and by what mode each was expressed by the Romans. Given an ut, he should know the full range of ideas possible for the speaker to have when he so begins a clause, and by what construction each of these ideas is expressed. And in particular it will be found useful to set before the class the whole range of verbal constructions that are capable of serving as the object or the subject or a verb (substantive clauses), and to ask them which and how many of these a given verb or phrase may take. These substantive clauses are as follows: —

• The indirect statement of fact (infinitive).
• The indirect question of fact.
• The indirect deliberative question.
• The final clause.
• The consecutive clause.

Now give the class a verb, dicit, and ask what possible completing verbal ideas there may be, and what phase of meaning one and another of these would indicate for the word dicit itself. The answer should be: the infinitive, if dicit means that a statement is made; the subjunctive introduced by an interrogative (including of course ut), if dicit means the giving of an answer to a question of fact or a deliberate question; the subjunctive with ut or ne, if dicit means the giving of a direction. The substantive consecutive clause, it is, of course, impossible for dicit to take. On the other hand, the meaning of a word like effecit is such that it can take the substantive consecutive clause and can take no other; so that, unless we find a clear accusative object, we are sure, upon meeting an efficit, that a verbal object introduced by ut or ut non is sooner or later to come. A verb like peto can take only a substantive final clause, a verb like quaero only an interrogative substantive clause (either a question of fact, or a deliberative question), etc. To look at these matters in this particular way is of great usefulness. If, for example, the clas is translating at hearing, in Cat. Mai. 63, the anecdote beginning quin etiam memoriae proditum est, everybody should at this point instantly recognize that an infinitive of statement is sooner or later inevitable, and, knowing the Latin habit of arrangement, should at once associate with that impending infinitive statement all the intervening matter, cum Athenis ludis quidam in theatrum, etc. The same thing is seen, with a much briefer suspense, in Caesar's “id si fieret, intellegebat magno periculo, etc.,” B.G. 1.10.2.