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

The sketch here given for the treatment of the accusative constructions suggests the way in which any set of constructions should be managed. I append a few specimens of the results for this and that class of words in a number of cases. I grant that the enumeration for the ablative, and even for the genitive, is of provoking sweep; but this is only equivalent to saying that the number of meanings of the ablative and genitive cases which a young student must learn, under whatsoever method, is great.

The genitive of any pronoun may be found to mean the possessor of some thing or of some activity (the activity being expressed in a verbal noun), or the object of some activity (expressed in a noun, an adjective, or some one of a certain list of verbs), or the whole of which some other word expresses a part, — may be, then, either subjective, or objective, or partitive; or it may simply belong to some noun, just as an adjective does. The genitive of any noun (say civitatis) may prove to be either subjective, or objective, or partitive, or in apposition with some other genitive. The genitive of a noun like periculi may prove to be either appositive, or subjective, or objective, or partitive, or (if modified by a noun or participle) qualitative. The genitive of a noun indicating an act or mental state of a bad nature may be either appositive, or subjective, or objective, including a crime charged or a penalty adjudged, or may be partitive. A genitive magni may agree with a noun, or may mean the value of something.

The dative of any word may mean the person or thing indirectly concerned in an act or state expressed by a noun or an adjective or a group of words. The dative of the name of a person (say Caesari) may have this general meaning, or, in one or another special phase of it, may mean the person concerned in an obligation indicated by a gerundive (the agent), or the possessor of something. The dative of a word like dolori, laudi, etc., may mean, in a general way, the thing indirectly concerned, or, with a special phase of that idea, may mean the end served.

The accusative we have discussed already. The vocative takes care of itself, when the form is unmistakable.

The ablative is a case to be dreaded. In general, it should, like other cases, be cut up as little as possible. Something can be done by proceeding from the three ideas of the starting-point, the means, and the place (true ablative, instrumental, and locative), as in Dr. Leighton's table on p. 290, and the table on p. 254 of the Allen & Greenough Grammar; but the best intentions on the part of grammarians and teachers have not yet made the matter easy for the learner. The suggestions to be given here must go beyond these three divisions.

Nearly all ablatives can be absolute, or can depend upon a comparative, or on a word like dignus or contentus. Beside this, a proper name (say Caesare) may be in the ablative of source, after some word like genitus, though such a form of expression is naturally rare in the prose read before going to college. Of course such a word cannot be in the ablative of means (in the narrower sense), or of specification, or of time, or of degree of difference. A word like die, however, beside the general possibilities, may indicate time, or the degree of difference, a word like auro means or price, a word like capite description, etc. I shall not attempt here a complete list of suggestions. In general, in spite of the complexity of the uses of the ablative, the learner is less likely to go badly astray in dealing with this case in actual practice than in dealing with the genitive or the accusative.

One point not yet touched upon is of the gravest consequence. when a form occurs which may be in either of two cases, or even possibly in any one of three or four cases, the pupil should not allow himself to suppose that he knows the case, even if a probability presents itself at once. E.g., a student reading in B.G.1.3, and passing by ea (his rebus adducti et auctoritate Orgetorigis permoti constituerunt ea, quae, etc.), may easily suppose ea to be the object of constituerunt, instead of waiting until conviction of some kind is forced upon him by the remainder of the sentence; which conviction will prove to be that ea was the object, not of constituerunt, but of an infinitive which is not reached until the quae-clause is finished. The direction to the student should be: Have your eyes open, but keep in doubt as long as possible; in a word, THINK, and WAIT.