Bryn Mawr Classical Review

Reviewed by Peter G. McC. Brown, Trinity College, University of Oxford (peter.brown@trinity.ox.ac.uk)

Version at BMCR home site

Table of Contents

Adolf Primmer (1931-2011) was a wide-ranging scholar who published monographs and articles on authors as diverse as Cato, Cicero, Virgil, Ovid, Seneca, Origen and Augustine in addition to the publications on Menander, Plautus and Terence collected here. One of his interests from an early stage of his career was the structure of the plays of New Comedy and Roman Comedy, and it was an excellent idea to assemble all his publications on this subject in a single volume, together with those on other aspects of these plays (see the Table of Contents link above for details).

Primmer was interested in structural analysis of both the Latin comedies of Plautus and Terence and the Greek comedies on which they were based; those Greek comedies have not survived, but he believed (as many scholars have done) that it was possible to reconstruct them on the basis of close examination of the Latin texts combined with what we know of Menander's practice from the surviving remains of his plays. His analyses were the product of sustained thought and based on powerful intuitions about how the plays that he was examining or reconstructing would work in the theatre; among other things, he insisted that one of Menander's aims in constructing his comedies was to keep his audience guessing about what would happen next. He had thought about the stage settings of the plays and the off-stage movements of the characters, and he sometimes offered very imaginative suggestions about how a scene should be played on the stage. Any reader will find details to quibble over, and his reconstructions, even when they seem entirely reasonable, are not obviously the only ones possible. Nonetheless, anyone working on the plays in question must take them seriously, and many of his observations of detail are well worth paying attention to whether or not one is convinced by his larger-scale arguments.

In the case of Menander's plays, he found it illuminating to analyse their plots in terms of the three phases protasis-epitasis-katastrophé (as outlined by the fourth-century grammarians Euanthius and Donatus) and to examine the relationship between these phases and the five-act structure generally accepted to have been employed by the authors of Greek New Comedy. Discussion of the distribution of a tripartite plot over five acts of a play has been ongoing since the Renaissance, but application of the debate to the plays of Menander has only gradually become possible in the last hundred years or so with the rediscovery of substantial portions of some of his plays. Primmer is not the only scholar to have applied it to Menander, but it is perhaps not surprising that there continues to be disagreement both about the detailed analyses and about the helpfulness of the three-phase approach altogether: certainly the plays (1) start by presenting us with characters and situations and (2) then introduce complications which (3) are finally resolved; but not everyone finds it illuminating to point this out, and where one places the borderlines between the phases is a matter of interpretation – as Primmer was well aware.

One of Primmer's most notable contributions was his insistence that each phase (and also each act) must have more than one sub-phase or focus of interest; this was sometimes a factor in his positing scenes for the Greek originals which he supposed Plautus to have cut. In his last publication ('Akte und Spannung: Zur hellenistischen Theorie der Komödienstruktur bei Aelius Donatus', published in 2008) he claimed to be able to trace this analysis in terms of sub-phases back via Euanthius to a Hellenistic Greek scholar who may or may not have been Theophrastus. On this view, Euanthius' definition of the epitasis as incrementum processusque turbarum ac totius, ut ita dixerim, nodus erroris preserves the memory of a definition which distinguished three sub-phases of the epitasis, and the image in nodus is that of a hard knot in a piece of wood, the climax of all the confusions. Similarly the definition of the protasis as primus actus initiumque est dramatis, rather than simply repeating the same idea twice, turns out to distinguish separate sub-phases, with initium dramatis referring to the scenes at the end of the protasis that lead to the epitasis. This seems to me to involve a certain amount of special pleading, and I am not convinced that incrementum and processus refer to different stages or that initium dramatis refers to something later than primus actus. But that in itself does not invalidate Primmer's elaboration of the three-phase analysis.

In the case of Plautus, Primmer believed that the Latin author deserved more credit than he has often been given for employing rather different patterns in his recasting of plays originally written by Menander and other authors: not a fixed number of acts, but a well-planned succession of units of action, each consisting of a sequence of metrical blocks and (at least according to the 1984 monograph on Bacchides) marked off from one another by musical interludes played on the pipes. Primmer was arguing against the generally accepted view that the Latin playwrights wrote for continuous action (unlike their Greek models), and he was clearly right that there are points of rest in the Latin comedies, just as there sometimes are within the acts of a Menandrian comedy. He was also right to insist that the action of a play by Plautus does normally progress towards a goal, however much Plautus' elaboration of individual scenes may sometimes distract us from perceiving that. The standard metrical pattern that he detected for the Plautine 'acts' was a succession of blocks of (1) spoken senarii, (2) cantica in a variety of metres, (3) trochaic septenarii. Exactly the same pattern has now been claimed to be standard for Roman Comedy by Timothy J. Moore, Music in Roman Comedy (Cambridge 2012), pp. 253-5, without reference to Primmer's analyses and without his belief in regular musical interludes. Moore allows that there are many exceptions to the pattern but claims they are often there for effect; Primmer allowed that the pattern was not strictly observed in the cases he examined but treated the passages that disrupted the pattern as minor distractions.

The plays Primmer discussed in most detail were Bacchides, Menaechmi and Aulularia, reconstructing their Greek originals not only by structural analysis but also by acute observation of the details of the text. Thus for Menaechmi it is a detail in the text of line 739 (a well-known problem, though Gratwick in his commentary contents himself with 'strictly, Ma. has no evidence for that') that leads him to argue that Plautus has abbreviated the scene in which it comes, with consequences for the presentation of the visiting brother in the scenes that follow; and his interpretation of resecro at Aulularia 684 as 'I formally retract' leads him to posit a scene earlier in the play in which Lyconides begged his mother to say nothing to his uncle about his love for Euclio's daughter. The reconstructions are ingenious, but I do not have the space to discuss them more fully. For Bacchides I think the reviews of his monograph give a fair idea of Primmer's main points and of the characteristics of his approach and offer (for the most part) reasonable criticisms. ¹

I end with brief remarks on the remaining papers. 'Zwei Terenz-Prologe' (1965) surveys the structure of the prologues to Andria and Heautontimorumenos, for the latter building on the more detailed discussion in 'Zum Prolog des Heautontimorumenos' (1964), which defends the coherence of the structure of this prologue and argues that Leo's analysis of it was too mechanical in the application of rhetorical terminology; notable is the suggestion, based on the confident tone of its prologue, that Heautontimorumenos must have been put on after the highly successful Eunuchus, not in the wake of the failure of Hecyra. In 'Die homo-sum-Szene im Heautontimorumenos' (1966) Primmer argues that HT 82 si quid laborist nollem means 'If our conversation is in any way upsetting for you, I wouldn't wish to – [sc. 'press you any further', omitted by aposiopesis]'. 'Zur Lektüre von Terenz' Eunuchus' (1979) is a survey of Eunuchus addressed to an audience of schoolteachers, partly summarising the scholarship on the play from Jachmann in 1921 to Büchner in 1974, partly focussing in detail on the opening two scenes, and ending with a discussion of II.3, which Primmer uses as a springboard for some general remarks about the differences between the five-act structure of Menander's original play and the metrical blocks which already in 1979 he believed to be fundamental to the structure of Terence's play.

'Karion in den Epitrepontes' (1986) reconstructs the scenes in the play that included the cook Karion. The reconstruction is generally reasonable, and much of it is accepted by Furley in his 2009 edition. On a matter not directly concerned with these scenes, Primmer's claim that Charisios must have appeared on stage in Act I seems difficult in the light of P.Oxy. 4936 (published in 2009) and was in any case not well supported by his argument from Menander's general practice; the idea seems to have been quietly omitted from the chart of the structure of this play on p.324 (published in 2008 already). On the other hand, his reordering of the fragments of P.Oxy. 3532 at p. 49 n. 3 was almost immediately confirmed by the publication of parts of the Michigan papyrus.

In 'Der Rudens bei Plautus und Diphilos' (2003) Primmer builds on Drexler's insistence that Daemones' summary of his dream at Rudens 593-610 offers important clues to the way the action developed in Diphilus' original play. His reconstruction of Diphilus' play seems reasonable, though his view that Daemones is a misanthrope cured of his misanthropy by the rediscovery of his daughter receives scant support from Plautus' text.

The volume has been prepared with care, starting with a helpful and sympathetic introduction and rounded off with a brief biography of Primmer and a full list of his publications. Marginal numbers indicate the pagination of the original publications. Errors in those publications have been silently corrected; of those that remain, the following are perhaps worth mentioning: p. 95 lines 4 and 6, 'Antipho' for 'Chaerea'; p.112 line 26, 'Pistoclerus' for 'Mnesilochus'; p. 167 line 9, 'flocked' for 'flogged'; p. 201 n. 40, '418' for '481'; p. 184, marginal number, '99' for '115'; p. 247 line 17, 'IV 2a' for 'IV 1'; p. 252 line 14, 'II 1-3' for 'III 1-3'; p. 295 n.1, 'Anm. 5' for 'Anm. 50'; p.314 n.61, 'Simo' for 'Laches'.

1.   Alain Blanchard, RÉG 100 (1987), 462-70; Ludwig Braun, Anzeiger für die Altertumswissenschaft 41 (1988), 161-3; François Jouan, Latomus 51 (1992), 186-8; Douglas N. Lacey, CJ 81.3 (1986), 262- 3; Eckard Lefèvre, Gnomon 57.8 (1985), 693-8; J.C.B. Lowe, CR 35.2 (1985), 396-7.