The subject of the infinitive

By socratidion, in 'Latin Grammar Questions', Jun 18, 2011.

  1. socratidion Civis Illustris

    • Civis Illustris
    This from Shelmerdine (on which see my initial post viewtopic.php?f=6&t=11643): it's from chapter 21, and all the emphasis is hers.
    This is simply incorrect. Shockingly, reprehensibly incorrect. Historic infinitives have a subject in the nominative, and even if Shelmerdine somehow thinks that doesn't count, there are plenty of other kinds of exception.

    Luckily for me (a total coincidence, honestly!) this very subject has been fully discussed in a thread posted earlier today: viewtopic.php?f=2&t=11748, so I have no need to go into detail about those exceptions.

    It's clear from that discussion, I hope, that infinitives have a nominative subject in sentences such as
    Hercules fortis esse vult/videtur/dicitur = Hercules wants/seems/is said to be strong. Here's one from Livy where the nominative is visible in the infinitive itself:
    Non ultra uidebatur latura plebes dilationem agrariae legis (Livy 2.63) = The plebs seemed not to be likely to put up with any further delay of the agrarian law.

    The subject of the infinitive can also be in the dative:

    atqui licet esse beatis (Hor. Sat. 1.1.19) = but it's possible for them to be happy.

    I haven't been able to find any examples of the subject being in the genitive or ablative, but as far as the infinitive is concerned, there would be no objection. If that seems paradoxical, it's probably because of the term 'subject of the infinitive' itself, which I detest -- seems a kind of contradiction in terms, like 'the god of the atheists'. Can I use the term 'word that controls the infinitive'? So in 'volo te venire', you would say that te 'controls' venire. Contrast Shelmerdine's terminology, where the object is the subject of the objective infinitive. Ugh. And I have to teach this stuff.

    So let's try this out as a rule: in Latin the infinitive doesn't care about the case of the word which controls it, but all other things being equal -- ie if there isn't a strong reason for it to be in any particular case -- it will tend to default to the accusative.

    This defaulting to the accusative is unsurprising: the other cases are too specialised (the nominative, for example, is too urgently the subject of the main verb). And of course in actual practice we see an accusative controlling the infinitive the vast majority of the time, so it feels normal, natural. So normal that we end up with things like 'te esse tristem me facit gaudere' (you being sad makes me happy), where te is used by default instead of tu because, logically 'you' is not the subject of 'facit'.

    Anyway, that was Shelmerdine's worst crime, and here the prosecution rests.
  2. Bitmap Civis Illustris

    • Civis Illustris
    Cygnea, Gena
    As I understand it, the book you're discussing is a textbook called "Introduction to Latin", not a comprehensive grammar. I think this is a good rule of thumb to work with when introducing Latin to a beginner. It may even be didactically questionable to overload a student with too much information at such an early point of study, where things like the "historic infinitive" would simply be confusing to him or her.

    The rule is not incorrect, it's simply incomplete. A book or a series with a good didactical conception would probably expand it later on saying something like "We have learnt that the subject of the infinitive is usually in the accusative case; however, there are also examples in which the infinitive may take a nominative as its subject" and then introduce the NcI for instance.

    I wouldn't criticise such points too much if they were made with the motive of didactical reduction.

    Yes, but the way such infinitives + dative come about is via some kind of analogy rather than a natural construction. They only occur with verbs like licere or necesse est which can have an AcI as their subject and a dative as their beneficial object. e.g.

    AcI necesse est iustitiam fieri
    Dat. id facio, quod mihi necesse est
    When the construction is joined with both, an infinitive and a dative object and the dative object is identical with the subject of the infinitive, the infinitive may attract that case, so you can say both hominibus necesse est esse beatos and beatis. This kind of case attraction also occurs elsewhere, for instance in mihi nomen est Marcus vs. Marco, both of which is acceptable.

    This is not so much an achievement of the infinitive as the influence of a general kind of analogy which can be found in the Latin language. As such, a dative as the infinitive's subject is more like an exception to the rule mentioned above (just like there are exceptions to the general rule that Latin word order is free) and there is no instance where an infinitive could only take a dative subject while the accusative would not be permissible (with the nominative, there is).

    The only verbs I can think of with a non-finite phrase and a genitive or ablative as their beneficial objects are refert and interest. However, the analogy described above does not seem to apply to those verbs. You can only say me[a-long:156bu5ig][/a-long:156bu5ig]/Ciceronis interest/refert esse beatum, but not *beati. Similarly, you can only say meum/Ciceronis nomen est Marcus, but not *Marci.
    So the objection you could have would be that Latin only allows case attraction for the dative, but not for the genitive or the ablative.

    I don't really see the problem. The "subject of the infinitive" is what the nominative subject of the sentence would be if it were not an infinitive construction. It makes sense to me.
    Pacis puella likes this.
  3. socratidion Civis Illustris

    • Civis Illustris
    It's not didactic reduction, it's didactic distortion. It's exactly this sort of thing that you find sticks in people's heads and they can't shift it for years. For heaven's sake, she emphasises it with bold type!
    It's just poorly formulated and therefore misleading, and it could so easily have been done better. The nominative+infinitive is just as basic, so it's not a refinement that can be addressed later. Even to introduce the 'rule' is itself a complication, when the constructions she is discussing are so intuitive anyway: volo te abire, that sort of thing.

    Secondly, the rule is bogus anyway: the infinitive doesn't have a grammatical subject, by definition, which is why a number of cases are capable of fulfilling an analogous role. That the accusative came to be the default is the result of a preponderance of instances of the 'volo te abire' type, and says nothing about what case the infinitive really 'requires'. It requires none...

    ...and so on -- I said this all in my first post, and I am uncertain why it bothers you now. You can argue it's better to simplify for didactic purposes: I agree, but I don't believe in telling falsehoods. You can argue that the infinitive has a subject which is in the accusative case as a rule, with exceptions; there you are merely contradicting me, which you may do, but it leaves me with very little I can respond to.
  4. Bitmap Civis Illustris

    • Civis Illustris
    Cygnea, Gena
    That sounds like you don't want anyone to reply to this :>

    I saw the link to this topic in some other post in the L-E forum. I've had thoughts on this matter before, but I never found the time to write them down.

    I'm just trying to find points in favour of your book; it would be boring if everyone simply agreed with you there :> The difference between didactic reduction and distortion can be pretty small at times. The rule may be falsely presented with the bold emphases and the questionable use of the word "subject", but the idea to restrain yourself to the use of the infinitive with the accusative case and deal with other cases only later is not fundamentally wrong ... although I doubt that this is what she actually had in mind.

    Then again, the whole idea of categorising the infinitive like that may not really be suitable for a textbook, anyway, because - as you say - the constructions are either very intuitive, anyway, or they are too specific (historic infinitive) to fit into a general scheme. When I learnt Latin, I simply learnt infinitive constructions 1 by 1, i.e. the AcI first, then NcI, then other things, without any general rules regarding the infinitive that would only have made it more confusing.
    I think if you consider her formulation to be wrong, incomplete or based on incorrect definitions, you should simply bypass it and leave it out altogether rather than trying to correct or expand it, which only makes it even more confusing for a beginner.

    Actually, the main point of my previous post was not really my attempt to defend Shelmerdine's integrity, but my thoughts on why the infinitive can only be governed by a dative (apart from the obvious nominative or accusative), but not a genitive or an ablative.

    It's because I'm not unfamiliar with this kind of diction. Over here, we usually speak about "accusative subjects" in AcIs. You have to be able to make some distinction when the infinitive is surrounded by 2 different accusatives, e.g. in "volo te mihi librum dare" - how do you do that if you don't use the terms "subject" and "object"?
  5. Manus Correctrix QVAE CORRIGIT

    • Civis Illustris
    Nothing wrong with talking about subjects.

    Shelmerdine should have said ‘tends to be in the accusative’.
  6. socratidion Civis Illustris

    • Civis Illustris
    As for how I actually dealt with it, I told my students to cross out 'is' and write 'can be'. Which I thought was elegant: both true, and addressing precisely the problem faced at this point in the book, where we are seeing accusatives as subjects of infinitive clauses for the first time.

    Trust me, I do not trouble my students with complications they don't need. But some of them are intelligent enough to follow the logic of these things to their conclusion, so accuracy is important even at this stage.

    As for my dislike of the term 'subject of the infinitive', it's a pragmatic thing, based on years of trying to teach students at beginners level, having to hammer into their heads "the SUBJECT is in the NOMINATIVE, and the OBJECT is in the ACCUSATIVE". Which is why I came up with 'the word that controls the infinitive' -- an intuitive description that doesn't sound too technical. My students get that.

    Though when I think about it, one doesn't say things like 'the object of a noun is in the genitive case': one recognises that such a formulation would be inappropriate for nouns.
  7. Nikolaos schmikolaos

    • Censor
    What about verbs that take a genitive object, though, like meminisse? Do you consider that indirect, even though it doesn't often make any difference whether it's genitive or accusative? And then, there are the objective genitive pronouns.

Share This Page


Our Latin forum is a community for discussion of all topics relating to Latin language, ancient and medieval world.

Latin Boards on this Forum:

English to Latin, Latin to English translation, general Latin language, Latin grammar, Latine loquere, ancient and medieval world links.