George Hinge


Epistolary deixis in the correspondence
of Fronto and Marcus Aurelius



Fronto’s style

Marcus Cornelius Fronto was the most influential rhetorician in second-century Rome. Born ca. 95 AD in Cirta in the Roman province of Africa, he studied in Rome and eventually became the tutor of the Emperor’s adopted sons Marcus Aurelius and Lucius Verus, who became emperors themselves later.

Unfortunately, the speeches of Fronto have not been preserved (except for shorter quotations), but a collection of his letters was re-discovered in a palimpsest manuscript by Cardinal Angelo Mai in 1815. This sensational found was, however, a disappointment.

On account of the Late Antique laudations, one expected a master of language and style who would match Cicero. Thus, Eumenius calls him, one hundred years after his death, Romanae eloquentiae non secundum sed alterum decus (Pan Lat. 8.14.2), i.e. Fronto came in, not in a second place, but in a shared first place in Roman eloquence. Similarly, Jerome mentions Fronto on the same line as Quintilian, Cicero and Pliny (Ep. 125.12.1).

The new-found letters gave a totally different impression. Here, we meet a hypochondriac philologist speaking about either linguistic quibbles or his own ever sickly health. Fronto’s unconditioned declarations of love addressed to the young Marcus Aurelius and this pupil’s equally unconditioned reciprocation of those declarations rendered the nineteenth-century readers as embarrassed as if one had entered a room and accidentally found a couple in an intimate embrace. Moreover, Fronto's style was considered mannerist and stilted. It was far away from the simple ideals of the epistolary genre.

In the classic monograph of Hermann Peter from 1901, Der Brief in der römischen Litteratur, Fronto is judged most severely: It is said, that he is “incapable of new creative thoughts and nevertheless diligent in his vanity to strike and attract the eyes upon him ... The pedantic orator had no understanding of the singular charm which natural letters show in the immediacy of the expression of sensations.” (pp. 124-125: “... unfähig zu neuen schöpferischen Gedanken und doch in seiner Eitelkeit bemüht aufzufallen und die Augen auf sich zu ziehen... Für den eigentümlichen Reiz, den die Unmittelbarkeit des Ausdruckes der Empfindung in natürlichen Briefen ausübt, namentlich in denen Ciceros an Atticus, besaß der pedantische Rhetor kein Verständnis”). The principal thesis of this book is that letters must be natural. In the recent Brill commentary to the letters of Fronto, Michel van den Hout is equally reserved, describing Fronto as a philologist – which is not meant as praise – and simply as “a third-class writer” (p. x.). One must say, there is some way from a shared first place to being a third-class writer!

Is it really true that the most important Latin orator of the second century had such a bad feeling of style? I doubt it: Modern critics base their judgment on anachronistic and false criteria, and in their monomaniac fondness of the natural – a romantic phantom – they fail to appreciate the skilled epistolary rhetoric practices in the correspondence of Fronto.


The Brill commentary to the letters of Fronto is both very erudite and very extensive, even though it is virtually impossible to use if one does not have Van den Hout's own Teubner edition on the desk. In spite of its flaws, the commentary is very useful when it comes to the stylistic analysis of the text, especially with its rich indices. It has a “Grammatical and Stylistic Index” (pp. 635-646), in which the author lists all known examples of particular linguistic phenomena or rhetorical figures in the letters. If one is allowed to assume that the selection of Van den Hout is representative (i.e. the instances that he must have overlooked are distributed equally in the corpus), one is able to draw some interesting conclusions about the distribution of these linguistic and rhetorical particularities. On the left page on the last sheet of the handout, I have presented some calculations made on the basis of Van den Hout’s references.

I have restricted the statistics to the five books ad Marcum Aurelium et invicem and the four books ad Antoninum imperatorem et invicem, i.e. the first 106 pages of the Teubner edition. Here, Van den Hout enumerates 27 ”archaisms” in Fronto’s own letters, but only 12 in the letters of Marcus Aurelius. The ratio is the opposite when one counts the so-called “colloquialisms”: Van den Hout has 52 in Fronto’s own letters, but 68 in the letters of Marcus Aurelius. Such figures are, of course, meaningless if they are not correlated to the extent of the letters of the two correspondents. If one takes the different extent into account, one may conclude that Fronto has 62.4% of the archaisms, but only 36.1% of the colloquialisms, whereas Marcus Aurelius has 27.7% of the archaisms, but as many as 63.9% of the colloquialisms. We see the same picture with other linguistic features as well: Several rhetorical-stylistic figures are found for the larger part in Fronto: e.g. alliteration 63%, chiasm 70%, metaphor 73%, homoioteleuton 73% and etymological figure 78% (all these numbers are ”calibrated” according to the different extent of the two correspondents). On the other hand, the linguistic features that are normally associated with a plainer style are much more frequent in Marcus Aurelius: post-Classical Latin 53% and diminutives 65%.

These calculations say nothing about the absolute location of Fronto and Marcus Aurelius in the stylistic spectrum of Latin as such. It would be a much more laborious calculation if one was going to compare, say, Fronto with Cicero. If such an examination should have any value at all, it should be same person who studied both corpora throughout, to be sure that the same principles were used in finding the various figures. Unfortunately, such an examination cannot be mechanised on the basis of electronic text collections, for the determination of these figures is based on both intelligence and judgment.

However, one can make other statistical investigations on the basis of an electronic text. Thus, the subjunction ut is more frequent in Cicero’s and Pliny’s letters than in Fronto and Marcus Aurelius, whereas the conjunction et is less frequent in the former than in the latter. One may infer that Cicero and Pliny is slightly more hypotactic and Fronto slightly more paratactic. You will find the numbers on the left page of the last sheet of the handout. I have also compared the frequency of the explicative conjunction enim with the causal subjunction quod. Unfortunately the latter number also includes the relative and interrogative pronoun quod – it would have taken me weeks to sort the different employments. In spite of the questionable methodological problems of such random calculations, it turns out, astonishingly enough, that Cicero, Pliny and Seneca show practically the same relative frequency of enim and quod; the same relation between these two words is found in Cicero’s speeches, whereas technical prose has numbers similar to those of Fronto and Marcus Aurelius. Even though it would be interesting to study these matters more thoroughly, it is not the subject of the present paper, which is dedicated to the deictic demonstratives.

The demonstrative pronouns hic, ille and iste are as a whole more frequent in the Ciceronian corpus, where they constitute 2.16% of all words – but only 1.87% in the correspondence of Fronto and Marcus Aurelius. In this light, it is interesting that iste is more than twice as frequent in the latter as in Cicero: 0.46% and 0.22%, respectively. This discrepancy calls for an explanation.

A priori, it may be seen in the context of a general tendency of the Latin language, according to which iste eventually replaces hic as the demonstrative pronoun with the meaning of ‘this’, a process, which has been completed in the Romance languages: French ce(t) and Italian questo go back to Latin iste.

One would find a confirmation of that possibility in the first letter of Marcus Aurelius (ad M. Aur. 1.2), where we find iste in juxtaposition with the possessive meus thrice. It would be conceivable that the more plain and modern style of Marcus Aurelius would include such a vulgarism. However, this is not confirmed by the relative frequency of the three pronouns in the letters of the two correspondents: hic, iste and ille are distributed practically identically. It is true that Marcus Aurelius uses a demonstrative pronoun relatively oftener than Fronto, which is probably a consequence of his more colloquial style. However, apparently, there was no functional difference in their choice between the particular pronouns.

I shall try to explain the predilection for iste otherwise, viz. with reference to the different deictic values of the demonstrative pronouns and the importance of deixis in epistolary discourse.


The title of my paper includes the word deixis, which is probably familiar to this audience already. However, I shall not abstain from recapitulating here what is meant with this technical term of pragmatic linguistics.

Fundamentally speaking, deixis, literally “demonstration” or “display”, is any linguistic means of locating an object, a person, an act or an utterance in space or time in relation to the speaker or her interlocutor. The most universal type of deixis is that of the personal pronouns and, somewhat less ubiquitously, the person inflection of the verbs. One may compare the sentence “George reads a paper to the audience” with “I’m reading this paper to you”. In the first case, the utterance is not related to the speaker or his hearer explicitly, and it is independent of time, space and situation.

The deictic system is centred on the speaker and her immediate physical and temporal surroundings – what is called the origo, “origin”, of the deixis – or, more accurately, the ego / hic / nunc -origo.

In most languages, different demonstrative pronouns are used to indicate various degrees of proximity to the speaker. English has a simple two-way referential distinction, this used for the things near the speaker physically or mentally and that used for things relatively distant from the speaker. This twofold system is the most frequent one in the human languages, as it is evident from the recent World Atlas of Language Structures (according to its census, a “two-way contrast” is found in 127 languages, a “three-way contrast” in 88 languages).

The second-most frequent is a tripartite distinction between proximal deixis, medial deixis and distal deixis, similar but not 100% equivalent, to the three persons of the personal pronouns and the verbal system. Thus, famously, Latin has the proximal pronoun hic corresponding to the first person, the medial pronoun iste corresponding to the second person, and the distal pronoun ille corresponding to the third person.

Most commonly, deixis is spatial or temporal – hic et nunc, time being derived from place metaphors. It is not, however, the only form of deixis. Thus, the so-called discourse deixis, or text deixis, is extremely common. E.g. in the sentence “And this taxing was first made when Cyrenius was governor of Syria”, ‘this’ does not indicate proximity to the author in time or space, but only in discourse, since it refers to the previous sentence. It is, however, the spatial deixis which is relevant here, because it plays an important role in the construction of the persons involved in the epistolary discourse

Deixis and genre

On the last page of the handout, you will find a statistic of the absolute and relative frequencies of the three deictic pronouns in different literary genres. The numbers derive from electronic texts which I have compiled for this purpose, and besides the relatively small correspondence of Fronto and Marcus Aurelius, I have tried to use rather large text corpora – more than 5,000 pages. I hope that this huge amount of text will compensate for the immanent inaccuracy of text string searching (rather than accumulating it).

The first column gives the number of instances of each pronoun, the second column their frequency in the text (percentage of the total amount words), and the third column gives their relative frequencies (percentage of the total amount of deictic pronouns).

The demonstrative pronouns together – the second column – are more frequent in some genres than in others, the extremes being historiography with 0.9% and comedy with 4.7%, i.e. more than five times as many. The frequent use of demonstrative pronouns is evidently characteristic of the colloquial language as said above. It renders the style more vivid, but at the same time also more context-bound and therefore less elevated. In English, too, the pronouns this and that are much more frequent in the colloquial language than in written prose, and I suspect that has similar stylistic implications.

Most instances of the deictic pronouns in historiography are examples of discourse deixis rather than spatial or temporal deixis. This explains the preference for the proximal pronoun, in the meaning of “the just-mentioned”. Especially common is the contrasting pair hicille, “the latter – the former”.

It leaps to the eye that the medial deictic pronoun iste is virtually absent from both historiography and technical handbooks. Thus, Caesar has it only once – significantly enough in a quotation of a speech (7.77.5). Similarly, I have found 86 examples of iste in all of Livy, and all of them occur in direct speech embedded in the narrative of his histories

On the other hand, iste is rather common in the speeches of Cicero and in the comedies of Plautus and Terence. It is natural that forensic oratory, which, no matter if it is defensive or accusatory, is directed against the other party in court, has plenty room for the so-called “second-person” deictic pronoun. Similarly, the high frequency of iste in comedy is not at all surprising, considering its dialogical nature.

Thus, it is only natural that iste is common in epistolary prose, too, since it is, after all, a written substitute of an oral dialogue, as it is often stated in both ancient and modern discussions. With this little word, the writer refers to the reader more or less subtly (more to us, less to the Romans) and with the alternating use of hic et iste, together with the personal pronouns and the person inflection, he pinpoints the mentioned persons, things and acts on a map defined by the writer and reader.

When considered isolated, iste is significantly more frequent in comedy than in all other genres. On the face of it, this fact would lead to the conclusion that iste was typical of the colloquial language, almost a vulgarism, which was avoided in the careful prose. It is, however, not the whole story. One has to take into consideration that all three deictic pronouns are much more frequent in comedy. So, it is not the predilection for iste that it is typical of spoken Latin, but rather the inclination to use demonstrative pronouns at all.

If we move from the second column of the table to the third one, we realise that, when considered relatively to the total amount of demonstratives, iste is in fact much more frequent in Fronto and Marcus Aurelius than in all other text corpora I have examined until now. It may be a coincidence due the limited size of their correspondence. However, I think it is worth trying to explain the discrepancy.

Deixis in Fronto and Marcus Aurelius

The epistolary discourse arises from a rhetorical embarrassment, viz. the actual absence of the interlocutor. The physical distance is of course bridged by committing the words to writing. However, the problem is not only physical, but also mental. Of course, if it were only about handing over important news or informations, one may be satisfied with a one-way communication. Most human communication is, however, not about reporting facts, but part of a social interplay (and this is true also for the types of communication which is about reporting facts).

So, epistolary discourse is first of all a virtual conversation taking place in a written medium. Each letter is written by one person, but a true conversation needs an interlocutor and not only a patient hearer. It is, therefore, essential that the writer stages a dialogue between the ‘I’ and the ‘you’ of the letter. Certain figures serve to attain this goal, e.g. apostrophes and imaginary questions and objections. Other epistolary common-places are explicit references to the friendship of reader and writer, previous and future encounters, the distance in time and space, and the difficulties in establishing a contact. All this contributes to draw a vivid picture of both persons involved in the virtual dialogue. This special epistolary dialogue is called epistolarity in the study of literature, and it rather trendy at the moment.

The over-frequency of the pronoun iste in both Fronto and Marcus Aurelius is probably part of this overall tendency of the epistolary discourse in establishing a common rhetorical space for the writer and the reader. This ought to hold true for the other epistolographic literature as well, but, as we have seen, iste is both absolutely and relatively less frequent in Cicero, Pliny and Seneca than in our correspondence. So, something special is going on here.

The over-frequency of iste is accompanied by an under-frequency of ille, whereas there are only minor differences in the choice of the proximal deictic pronoun hic. Therefore, if one is allowed to draw such inferences from the raw numbers alone, it looks like the correspondence of Fronto and Marcus Aurelius shuts the outside world, the ille-world, out. This is consistent with the over-all tone of the letters, which are characterised by a very intimate, almost amorous, note. Another – not alternative, but additional – explanation for this linguistic idiosyncrasy is perhaps a greater consciousness of the rhetorical expectations of the epistolary genre on behalf of both correspondents. After all, the letters were sent between a first-class rhetorician and his most skilled pupil, and what is more, the primary purpose of most of the letters is not practical, but exclusively the maintenance of the men’s special friendship.

ad M. Caesarem 1.2-3

But let’s look on some texts. In the handout, I have included a couple of letters displaying the dynamics of the deixis. In the Latin text of the handout, I have indicated the hic pronouns and adverbs with a double frame, the iste pronouns and adverbs with a solid frame and the ille pronouns and adverbs with a dotted frame. Furthermore, the deictic demonstratives are tangled with personal pronouns and verbal person inflection, which are shown in the text with a double underlining for the first person and a solid underlining for the second person. Even though the result may seem chaotic at first sight, I hope, it gives a good impression of the vivid interaction between deixis and person in the text of the letters.

In the first letter on the handout, 1.2, Marcus Aurelius complains that he is not able to run to Fronto and nurse his sick foot. It is full of medial deictic demonstrative pronouns and adverbs. As I mentioned above, Marcus Aurelius combines the second-person demonstrative iste with the first-person possessive meus three times in this letter. Such combination is rather rare in Classical Latin, and Michel van den Hout thinks that the pronoun iste here indicates something unpleasant. The derogatory semantics of iste is well-known in the speeches of Cicero. There is, however, an obvious difference between the rhetorical context of those forensic texts and these letters – above all in the value of the second-person. It must be born in mind that the calamities qualified by the demonstrative iste are suffered by the writer, but they are at the same time all related to the addressee, as far as they are caused by the inability of the writer to take proper care of the addressee.

The two instances of hic in this letter are examples of discourse deixis, whereas ille refers to Fronto twice. The choice of demonstrative underlines the remoteness of the addressee. At the same time, the displacement of iste into the first person and ille into the second person emphasises the dynamics of epistolary discourse.


Fronto’s answer, 1.3, is a rhetorical masterpiece, both a skilled play with the epistolographical topics and a sophistic defence of pure friendship.

It is arranged beautifully like a speech with the five traditional divisions: exordium, narratio, propositio, argumentatio and conclusio. Such a division is normally not found in epistolary prose, not even in Fronto, because it breaks the illusion of a free conversation. Thus, this contribution presents itself, in its external form, as a pompous monologue, but this impression is softened by the constant references to the addressee and the situation of the letter, thus in the vivid description of the messenger waiting impatiently for Fronto’s letter towards the end (already introduced in §2).

The first sentence of the letter touches an iambic senar without being one, like it is typical in the metrical allusions in prose literature: Tu Cáesar Frón|ton(em) ístum tuúm | sine fín(e) amás. The proem arouses the interest of the reader immediately since it declares, not the love of the writer to the addressee, but the love of the addressee to the writer. The point of departure is a paradox, and the whole letter must be analysed in the light of this paradox. The semi-iambic introduction underlines the teasing aspect of the letter, and at the same time, it marks the text off as a piece of colloquial literature and not a philosophical monologue.

The letter is full of iste pronouns, probably as a conscious response to the many instances of iste in the preceding letter of Marcus Aurelius. At the same time, it paints the discourse in bright second-person colours. It qualifies the immediate impression of the letter as a philosophical treaty on the true nature of love. On the contrary, it is epistolary discourse all the way, even though the second person fades momentarily in the stilted contrast of Reason and Chance in the second half of the letter. The true love and the images associated with it are determined by the demonstrative ille, thereby marking them off as a distant tableau, which is eventually adduced as an independent, and therefore more credible, argument for the very special nature of Fronto’s and Marcus’s friendship.

Most of the instances of iste refer unequivocally to Marcus Aurelius. In Haines’s translation, they are rendered with ‘your’. More complicated is Crede istud in §2, since the pronoun here refers to a condition inside Fronto himself. However, since this condition is more precisely caused by the previous letter, there may be some justification for the deixis. The monologic form is also modified by the many second-person pronouns (29x) and verbs (23x) – alternating with a similar number of first-person pronouns (21x) and verbs (33x). The fat that the first person dominates in the verbs and the second person dominates in the pronouns, gives the impression that the writer is an agent, whereas the addressee is a point of reference is this letter. This impression is consistent with the actual content of the letter.

ad M. Caesarem 2.2

In the third letter of the handout, 2.2, the first two paragraphs are dedicated to the proximal space with three instances of the pronoun hic, but the balance tips in favour of iste later in the letter. However, if one takes a closer look at the text printed on the handout, it is evident that the second person is present in the first paragraphs as well – and conversely, the first person is more frequent in the later paragraphs, where iste dominates:

Demonstrative pronouns

Personal pronouns

Verbal forms











































The medial deictic demonstrative occurs for the first time in the phrase Facies istud, somewhat surprisingly, since istud refers to what has been said. One would expect a discourse deictic hoc instead. However, it is a point in Fronto’s argumentation that this is already a virtue of Marcus. So, the choice of demonstrative emphasises this line of argumentation: “Act in this way characteristic of you.”

The following three instances of iste qualify two different literary authorities, Cicero and Horace. The choice of demonstrative is somewhat atypical. In the first case, Haines translates it ‘that’, whereas in the other two, he simply omits it. The first iste may be justified as a proper ‘second-person’ deixis, since it is, after all, Marcus who writes in that Ciceronian vein.

In all three examples, regular literary prose would have used the distal deixis pronoun ille just like that, since they refer to something which is definitely outside the physical space of the writer. However, the choice of demonstrative places the persons and matters loudly in the sphere of the addressee, given that the relatively rare iste is more marked than hic or ille. So, even though each single case may not be justified logically, the sum of them contributes to the epistolary discourse.

ad M. Caesarem 3.2-3

Marcus Aurelius writes 3.2 to Fronto in a diligent matter. Fronto is involved in a lawsuit against a client of Herodes Atticus, which happens to be Marcus Aurelius’ teacher in Greek rhetoric. It is foreseeable that Fronto will attack the patron as well in his speech, and Marcus Aurelius tries to calm his teacher, before the situation is getting too unpleasant for all parties. The letter has three deictic pronouns:

  • the first of them, hac, is used for discourse deixis;

  • the second of them, illum, designates adequately the third party, Herodes Atticus; and

  • the third of them, istud, qualifies the “tiresome case”, negotium odiossisimum, which F. is going to conduct

The three pronouns are lone-standing pillars in Marcus’s ironic rhetoric. The virtual dialogue is established with reference to a frequent request advanced by Fronto and an imaginary question put in his mouth. Furthermore, the argumentation is centred on a friendship topic. So, the discourse is clearly epistolary (rather than deliberative rhetoric).

Fronto’s response, 3.3, on the other hand, has more deictic pronouns, twice hic, four times iste, and thrice ille: The instances of hic and ille are examples of discourse deixis, but the four instances of iste are more complicated:

  • ista clearly refers to the second person, viz. Marcus’s dissembling irony

  • in isto negotio and in istis criminibus may be examples of the derogatory use of iste, typical of the forensic rhetoric, ~ “the accused”

  • iste Herodes may contain this forensic iste as well

On the other hand, this is not a forensic speech directed against the accused party in a trial. iste may be paraphrased “your” in the last case – and this is also the translation of Haines. Of course, this is not possible in the other two examples, for it would not make any sense to ascribe the case and the charges to Marcus. However, iste is not simply equivalent to tuus, ‘your’; it may refer to things connected with the second person indirectly also. If Herodes is iste Herodes, the charges against his client may also be called ista crimina.

Furthermore, in both instances, Fronto relies explicitly on Marcus’s letter and his advice. So, the negotium and the crimina belong to the preceding and future contributions to the epistolary dialogue. Thus, the choice of pronoun places the case in the framework of the epistolary exchange.

ad M. Caesarem 3.9

In the last letter given on the handout, there is a balance between the first and the second person in both verbs and personal pronouns:




verbal forms



personal pronouns






The virtual dialogue is also emphasised with the demand that the addressee writes back (scribe mihi), and that he sends him his new product (mississes mihi istud novicium), with references to previous remarks by the addressee (ais, minaris, concastigabas), and finally with an imaginary question put into the mouth of the addressee (“Quamobrem?” rogas).

The first instance of iste, istud novicium, may easily be paraphrased ‘your’; the other one, the adverb istīc in the following sentence is more difficult: Haines translates it ‘here’. The verbal act which it qualifies is in the first person and it is located in the home of writer. But why doesn’t he simply choose the proximal adverb hīc? In his commentary, Van den Hout refers to the first letter of Marcus where it is said that the meaning ‘here’ “occurs for the first time in our correspondence”. He mentions four other instances in the letters of Marcus (ad M. Caes. 2.8.3, 2.10.2, 2.11.2, 4.2.3) and one in the letters of Fronto (Laudes 1.2) and considers it a colloquialism, which is common in the later writers from Tertullian onwards.

However, in the first letter this improper istīc should not be isolated from the three instances of iste meus, which I analysed, not as derogatory, but as referring to the addressee, insofar as the worries that they qualify originate in the sufferings of the addressee. When, in the same context, it is said that the worries “keep me a prisoner istīc”, it is illogical, if one demands that the latter adverb must designate the space in which the addressee is located. But apparently, it may also designate a space caused by the addressee in some way or other. This analysis may also be true in 3.9: the writer reads the words of the addressee istīc, i.e. in a place and situation conditioned by the addressee, but not connected with him physically.

Or, one might analyse istīc in these two cases and in the other four in Marcus’s letters, not as a ‘here’ defined as the space surrounding the speaker (the origo), but as a ‘there’ defined from the point of view of the interlocutor. This analysis does not exclude the other analysis, but complements it. At the same time, the turning-upside-down of the origo contributes to the dynamics of the epistolary discourse, which attempts to establish a virtual dialogue, i.e. an exchange of shifting points of view.


Let’s sum up:

Iste is relatively more frequent in the correspondence of Fronto and Marcus Aurelius – and even more so when one takes into consideration that they have fewer demonstratives in general than most dialogic literature.

The predilection for this particular demonstrative pronoun is connected with the rhetorical demands of epistolary discourse, and this general tendency is accentuated even more in our correspondence due to the very intimate atmosphere of those letters. The outside world, normally represented by ille, is to a large extent shut out, and the scene is taken over by an exchange of hic and iste tableaux.