I was wondering about the correct analysis of the dative mihi in the sentence Magnum enim est in bonis praesidium quod mihi in perpetuum comparatum est, which is included in the text below from Cic. Cat. 3.12.27:

[27] sed quoniam earum rerum quas ego gessi non eadem est fortuna atque condicio quae illorum qui externa bella gesserunt, quod mihi cum eis vivendum est quos vici ac subegi, illi hostis aut interfectos aut oppressos reliquerunt, vestrum est, Quirites, si ceteris facta sua recte prosunt, mihi mea ne quando obsint providere. mentes enim hominum audacissimorum sceleratae ac nefariae ne vobis nocere possent ego providi, ne mihi noceant vestrum est providere. quamquam, Quirites, mihi quidem ipsi nihil ab istis iam noceri potest. Magnum enim est in bonis praesidium quod mihi in perpetuum comparatum est, magna in re publica dignitas quae me semper tacita defendet, magna vis conscientiae quam qui neglegunt, cum me violare volent, se indicabunt.

As I see it, the sentence at issue could in principle (i.e. grammatically speaking) have three possible readings depending on the kind of participle and dative involved:

  1. verbal/eventive participle + beneficiary dative: 'the protection {has been/was} secured for me for ever';

  2. adjectival/stative participle + the dative typically associated to this kind of construction (the so-/mis-called "dative of agent"); see also the typical parallelism between mihi est comparatum and habeo comparatum; 'I have a protection secured for ever';

  3. adjectival/stative participle + beneficiary dative: 'the protection is secured for me for ever'; see the translation below given by C.D. Yonge (Perseus).

[27] But since the fortune and condition of those exploits which I have performed is not the same with that of those men who have directed foreign wars—because I must live among those whom I have defeated and subdued, they have left their enemies either slain or crushed,—it is your business, O Romans, to take care, if their good deeds are a benefit to others, that mine shall never be an injury to me. For that the wicked and profligate designs of audacious men shall not be able to injure you, I have taken care; it is your business to take care that they do not injure me. Although, O Romans, no injury can be done to me by them, —for there is a great protection in the affection of all good men, which is procured for me for ever; there is great dignity in the republic, which will always silently defend me; there is great power in conscience, and those who neglect it when they desire to attack me will destroy themselves. [italics & bold mine: Mitomino]

When discussing some tricky interpretations of the dative in Latin (in particular, the one related to so-called dativus auctoris ("the dative of agent")), Guy Serbat, one of the most distinguished French Latinists of the second half of the XX century, wrote the following in his monumental "Grammaire Fondamentale du Latin", 1996, p. 502; NB: Engl. translation mine [Mitomino]:

Il est vrai que l’interprétation exacte du D[atif] est parfois délicate, surtout avec un verbe qui peut être suivi de la mention d’un bénéficiaire, comme emere (ci-dessus), ou comparare. Dans Cic., Cat. III 27, praesidium mihi comparatum est, faut-il comprendre «une protection m’a été procurée»; ou «la mise en place d’une défense est, pour moi, chose faite»? Le contexte montre que la deuxième interprétation est la bonne.

It is true that the exact interpretation of the D[ative] is sometimes tricky, especially with a verb which may be followed by the mention of a beneficiary, such as emere (above), or comparare. In Cic., Cat. III 27, praesidium mihi comparatum est, should we understand “a protection has been prepared for me” or “the preparation of a defense is, for me, done”? The context shows that the second interpretation is the correct one.

I've taken a look at some translations of this example from Cicero and I've realized that the translators disagree on the correct analysis given to this dative. Some consider it as a simple beneficiary (see Yonge's transl. above), whereas others appear to interpret it rather as a "dative of agent". Personally, I tend to agree more with G. Serbat's view rather than with C.D. Yonge's translation.

Serbat (1996: 502) also criticizes Kühner & Stegmann (II 1, 324-325) for whom datives like the one above are, at the same time, expressing the agent (Germ. «Urheber») and the beneficiary. Serbat (1996: 502) criticizes these authors as follows:

Il nous semble que c’est confondre le message linguistique et son interprétation dans une situation donnée (...) Les ambiguïtés que provoque une phrase isolée se ramènent donc à la question suivante: de quel morphème le D[atif] est-il «complément»? de -ndo- et de -to-, c’est-à-dire de «nécessité» et de «chose accomplie»? ou du thème lexical du verbe? [It seems to us that this is mixing the linguistic message and its interpretation in a given situation (...) The ambiguities caused by an isolated sentence therefore come down to the following question: Of which morpheme is D[ative] the "complement"? Of -ndo- and -to-, that is to say, of "necessity" and "done thing"? Or of the lexical stem of the verb?]

In my opinion, Serbat is right when saying that, grammatically speaking, we must distinguish two kinds of datives here and not mix them (according to him, the one related to the lexical base and the one related to necessity (-ndo-) or "chose faite" 'lit. done thing' (-to-)), i.e. Serbat is right when saying that in an example like mihi domino est parendum one must distinguish the argumental dative of the lexical verb (domino) and the so-/mis-called "dative of agent" (mihi) [NB: this example is in principle grammatically ambiguous: mihi could also be interpreted as the argumental dative, domino being then the "dative of agent"]). Ditto when dealing with examples like mihi est pugnandum imperatori, where one must distinguish the beneficiary dative (imperatori) from the so-/mis-called "dative of agent" (mihi) [NB: as above, this example is also ambiguous]. The "dative of agent" and the beneficiary dative are occupying different positions in the grammatical representation, the former being more prominent than (i.e. being hierarchically superior to) the latter.

In the comment section below, Sebastian Koppehel raises an interesting question: "Do we have unequivocal examples for the dativus auctoris where the agent is unequivocally not the beneficiary?". For cases involving the gerundive, the answer is positive (e.g. mihi est pugnandum imperatori) but it is true that it is not so obvious for cases involving perfect participles, basically for reasons related to the semantic nature of the lexical verbal predicate included in the stative participle (by far much more restricted than the gerundive), whereby the dissociation can be more difficult to pin down. But the relevant point is that a dissociation of these two nuances (beneficiary & agent) should be expected (see the following example drawn from Section 375 of A&G: Mihi rēs prōvīsa est. (Cic. Verr. 4.91) 'The matter has been provided for by me').

If Serbat is right (and I think he is; see above), one should not mix the grammar of a beneficiary dative with that of a so-/mis-called "dative of agent" (although it is true that some Romance translations I've consulted do make this mix. For example, a typical translation I've found is the following: '(a protection) that I have secured for myself for ever', where "I" captures the dative-of-agent nuance and "for myself" expresses the beneficiary nuance).

When dealing with the distinction between beneficiary dative and "dative of agent", Sebastian says: "Your Romance translations do indeed mix, but it is difficult to condemn them, as Cicero was undoubtedly the beneficiary". Yes, Sebastian is right, but the relevant point here is not translation to English/Romance languages but Latin grammar. Otherwise, note that, for example, the grammatical (!) difference between a beneficiary dative and a "dative of agent" referred to in the following paragraph (related to a previous post of mine) would not arise:

Although I agree with Serbat's criticism of Kühner & Stegmann, I disagree with him in his claim that the so-/mis-called dativus auctoris is directly related to the suffix -ndo- in gerundival constructions and to the suffix -to- in (stative) participial constructions. Otherwise, as I pointed out in this post, one could not explain why Hoc praesidium mihi comparatum est is ambiguous (both a beneficiary and "dative of agent" readings are possible), whereas Hoc praesidio comparato mihi is not ambiguous (only the beneficiary reading is possible in this non-verbal context). As pointed out in the post linked above, (true) "datives of agent" are not expected to appear in non-verbal contexts (e.g. in ablative absolutes or, more generally, in dominant participle constructions).

To conclude, following Serbat's and my remarks above on the (grammatical) need of separating the so-/mis-called "dative of agent" from the beneficiary one), I'd go (not without some hesitation: hence my question) for the following analysis: grammatically speaking, the dative at issue here is an instantiation of the so-/mis-called "dative of agent" (namely, the one involved in reading 2 above), although, contextually & conceptually speaking, there's nothing wrong in interpreting (and translating!) mihi as a beneficiary and agent (e.g. see above: '(a protection) that I have secured for myself for ever').

Please let me know if you have a different analysis/reading (see the three ones above) of this sentence. Thanks!

  • 1
    Yonge's interpretation seems correct to me, see the parallelism: magnum est praesidium, quod mihi comparatum est, magna dignitas, quae me defendet, etc. But what I'd like to know is what "context" supposedly shows that this is not right. Jul 22 at 20:30
  • @SebastianKoppehel Thanks for your comment. As I see it, the sentence has three possible readings: 1) eventive/verbal participle + beneficiary dative ('the protection {has been/was} procured for me for ever'); 2) adjectival participle + the dative typically associated to this kind of stative participle ('I have a protection procured for ever'; cf. the miscalled "dative of agent"); 3) stative/adjectival participle + beneficiary dative ('the protection is procured for me for ever'). I think Serbat is right when saying that, grammatically speaking, one cannot mix readings 2) & 3), as K&S do.
    – Mitomino
    Jul 23 at 0:12
  • This said, most of (Romance) translations I've checked make this kind of mix criticized by Serbat by translating the sentence as lit. '(a protection) that I have procured for myself for ever' (cf. Yonge's alternative translation above, which seems to be more compatible with reading 3 above). I don't know why Serbat criticizes the following interpretation: «une protection m’a été procurée», related to reading 1) above. Hence, in part, my question.
    – Mitomino
    Jul 23 at 0:22
  • It took me a while, but I think I now understand your point: If the dativus auctoris with comparatum is truly the, shall we say, perfect counterpart to the one with comparandum (primum praesidium mihi comparandum erat, comparavi, nunc mihi comparatum est), then where does the beneficiary aspect come from? I also now see why Serbat thinks "la deuxième interprétation est la bonne" : an impersonal passive with a beneficiary dative suggests someone else actively procured a protection for Cicero, but nobody did such a thing (for as we know, Cicero is the only person [of the good guys] who ... Jul 24 at 21:06
  • 1
    @tony «la mise en place d’une défense est, pour moi, chose faite» is just a semantic paraphrase used by Serbat to express HIS particular understanding of this dative. This paraphrase sounds very stilted/unnatural in French (and in English, as you rightly note), but this is not the point. The point is how one should understand this dative here: e.g. as a mere beneficiary dative (see Yonge's transl.) or as a "dative of agent", etc. Cf. the typical parallelism between Mihi est hoc comparatum & Habeo hoc comparatum;see also Section 375 dcc.dickinson.edu/grammar/latin/dative-agent
    – Mitomino
    Jul 25 at 15:50

1 Answer 1


The beneficiary reading seems obvious to me and interpreting mihi as an appositive, like meam vicem or something like that seems very strained to me and more of modern thought process than a Roman one. To make such an interpretation you would really need to prove that the Romans even used mihi in that way ever. When mihi is found in at the beginning of subordinate clauses like this, it universally has the usual meanings where it is auxiliary to the verb of the clause. In rare cases where a Roman wants to express an idea like "according to my way of thinking", specific verbiage is used to make that clear and that verbiage is set off from the rest of the sentence normally by adverb like sicut. You can't just throw in a mihi as a way of saying "as-for-me" the way you can in English.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.