Hostname: page-component-cd9895bd7-dzt6s Total loading time: 0 Render date: 2024-12-27T22:55:40.518Z Has data issue: false hasContentIssue false
  • English
  • Français

The Date of Polyperchon’s Invasion of Macedonia and Murder of Heracles

Published online by Cambridge University Press:  08 May 2015

P.V. Wheatley
P.V. Wheatley*
Affiliation:
University of Western Australia
Get access Rights & Permissions [Opens in a new window]

Extract

The status and activities of Polyperchon in the period following 315 B.C. remain obscure to scholars, even after the several detailed reconstructions of the Diadochian era that were made during the 1980s. Heckel refers to him as a ‘jackal among lions’, Hornblower as one of the ‘second-class’ generals. He was propelled to prominence upon inheriting Antipater’s mantle as guardian of the young kings and supreme commander in late 319 (Diod. 18.48.4), and had ample opportunity to make himself the dominant figure among Alexander the Great’s Successors in Europe. However, he was indecisive, unlucky, and more than a little incompetent in his handling of affairs in Greece and Macedonia over the next two years, being thoroughly outmanoeuvred on the military, diplomatic, and propaganda fronts by Antipater’s son, Cassander. By 316 Polyperchon had lost custody of the royal family, and was little more than a fugitive in Aetolia, but ambiguity regarding his legal and diplomatic status in relation to the regency and the other Diadochs continued to dog him until his disappearance from recorded history some fifteen years later.

Type
Research Article
Information
Antichthon , Volume 32 , November 1998 , pp. 12 - 23
Copyright
Copyright © Australasian Society for Classical Studies 1998

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 E.g. Billows, R.A., Antigonus the One-eyed and the Creation of the Hellenistic State (Berkeley 1990)Google Scholar; Grainger, J.D., Seleukos Nikator: Constructing a Hellenistic Kingdom (London 1990)Google Scholar; Lund, H.S., Lysimachus: a Study in Early Hellenistic Kingship (London 1992).CrossRefGoogle Scholar

2 Heckel, W., The Marshals of Alexander’s Empire (London 1992)188Google Scholar; Hornblower, J., Hieronymus of Cardia (Oxford 1981)224Google Scholar. On Polyperchon in general, see Lenschau, T., RE 21.2 (1952) col. 1798–1806Google Scholar, no. 1; Berve, H., Das Alexanderreich auf prosopographischer Grundlage 2 (Munich 1926), no. 654Google Scholar; Heckel 188-204.

3 Diod. 19.52.6. For detailed analysis of the events of 319-316 concerning Polyperchon, see Heckel (n.2) 193-201.

4 For two opposing views regarding the legitimacy of his claim, see Tarn, W.W., ‘Heracles Son of Barsine’, JHS 41 (1921) 18–28CrossRefGoogle Scholar, contra Brunt, P.A., ‘Alexander, Barsine and Heracles’, RFIC 103 (1975) 22–34Google Scholar, whose exposition is compelling.

5 Seibert, J., Untersuchungen zur Geschichte Ptolemaios’ I.’ (Munich 1969)186Google Scholar; Das Zeitalter der Diadochen (Darmstadt 1983) 129Google Scholar; Bakhuizen, S.C., Salganeus and the Fortifications on its Mountains (Groningen 1970) 125, 162–164Google Scholar; Adams, W.L., Cassander: the Policy of Coalition (diss., U. of Virginia 1975) 146Google Scholar; Brunt (n.4) 34; Mendels, D., ‘Aetolia 331-301: Frustration, Political Power, and Survival’, Historia 33 (1984)176Google Scholar; Hammond, N.G.L. & Walbank, F.W., A History of Macedonia 3 (Oxford 1988) 169Google Scholar; Billows (n.l) 140-141; Heckel (n.2) 203; Lund (n.l) 65; Carney, E D., ‘Olympias, Adea Eurydice, and the end of the Argead Dynasty’, in Ventures into Greek History, ed. 1. Worthington (Oxford 1994)357–380Google Scholar. esp. 377 (‘late 309’). See also the discussion of Bosworth, A.B. in ‘The Political Context of the Liber de Morte’, now entitled ‘Ptolemy and the Will of Alexander’, presented at the ‘Alexander the Great: History and Romance’ Symposium at the University of Newcastle in July 1997Google Scholar (proceedings forthcoming, 1999).

6 E.g. Will, E., Histoire politique du monde hellénistique (323-30Google Scholar av. J, C.) 1 (Nancy 1966) 60, 63Google Scholar; Bengtson, H., Die Strategie in der Hellenistischen Zeit 1 (Munich 1937, 1964) 136–137Google Scholar n3; Smith, S., ‘The Chronology of Philip Arrhidaeus, Antigonus and Alexander IV, Revue d’Assyriologie 22 (1925) 195Google Scholar; Tarn, W.W., CAH 6 (1953) 493Google Scholar n.2. Lenschau, , RE 21.2 (1952)Google Scholar col. 1805 is also aware of the possibility. The best treatment of the question is Bakhuizen (n.5), app. IX, 162-164, who emphatically dates Polyperchon’s invasion of Macedonia to 309, and includes bibliography. His discussion is not conclusive, though, as I shall show.

7 Diod. 19. 53.1; 57.5; 60.1; 61.1; 62.5; 63.1; 63.3.

8 Diod. 19.64.3. For discussion, see Bakhuizen (n.5) 121 n.55; Mendels (n.5) 165 ff., which is only flawed by the adoption of the erroneous ‘low’ chronology for the period. For the ‘high’ chronology presumed by this paper, see Wheatley, P.V., ‘The Chronology of the Third Diadoch War, 315–311 B.C.’, Phoenix 52 (1998Google Scholar, forthcoming), contra Errington, R.M., ‘Diodorus Siculus and the Chronology of the Early Diadochoi 320-311 B.C.’, Hermes 105 (1977) 49–77.Google Scholar

9 Diod. 19.67.1-2 (Alexander’s assassination and Cratesipolis’ succession); 19.74.2, 20.20 (Polyperchon’s position, 313-310). On Alexander, see Berve 2 (n.2) no. 39. On Cratesipolis, see Macurdy, G.H., ‘The Political Activities and the Name of Cratesipolis’, AJP 50 (1929) 273–278.Google Scholar

10 Simpson, R.H., ‘Antigonus, Polyperchon and the Macedonian Regency’, Historia 6 (1957) 371–373Google Scholar discusses the apparent transfer of the regency from Polyperchon to Antigonus in 315, concluding that this was not within Polyperchon’s legal power, but that relations between them were certainly friendly thereafter. From 315-312 Antigonus sent at least three of his trusted lieutenants to the Péloponnèse: Aristodemus in 315 (Diod. 19.57.5); Telesphorus in 313 (Diod. 19.74.1-2); and Polemaeus in 312 (Diod. 19.87). It is unclear how these men related to Polyperchon and his allies. Telesphorus operated against Alexander’s garrisons in 313, but left Polyperchon alone in Corinth and Sicyon, while Polemaeus was concerned only with Telesphorus in 312. Bakhuizen (n.5) 112 ff. provides the most detailed discussion of these events, esp. p.l 17, and notes 22, 28 and 55. His remarks hypothesising that Polemaeus’ revolt against Antigonus in 310 was a direct result of the latter’s connivance with Polyperchon are most persuasive, as is the observation (n.55): ‘It may be significant that Polemaeus was in the Peloponnesos when he revolted’. See also Adams (n.5) 112 ff.; Hammond (n.5) 157 ff.

11 OGIS 5, 11. 37–40Google Scholar = Welles, C. B., Royal Correspondence in the Hellenistic Period: a Study in Greek Epigraphy (New Haven 1934)Google Scholar, no. 1. Polyperchon seems to have retained some semi-independent status, although technically in a loose alliance with Antigonus against Cassander.

12 Diod. 20.19.2, recording Polemaeus’ defection before Polyperchon’s military buildup. This counts against the theory that Polemaeus’ treachery was partly due to Antigonus’ cooperation with Polyperchon, but I think Diodorus’ sequence cannot be pressed too hard. The entry describing the old general’s preparations (20.21) is certainly retrospective in tone and context, giving the impression that he had been brooding and planning for some years, but it was not until the murder of Alexander IV that the opportunity for a coup presented itself. On the spelling of Polemaeus’ name, which Diodorus renders πτολΐμαίος (cf. Plut. Eum. 10.3), causing confusion with the Egyptian Ptolemies, see Hauben, H., ‘Who is Who in Antigonus’ Letter to the Scepsians’, EA 9 (1987) 29–36Google Scholar, esp. 31-32; and Billows (n.l) 427, no. 100. For the purposes of this article, I have used the form found in the primary evidence: πολΐμαιος (IG ii 469, line 4; IGSK 28.1 no. 2, lines 2, 6,10,11).

13 Diod 19.105.2-4; Marble, Parian (FGrH 239) F B18Google Scholar; Trogus, , Prol. 15Google Scholar; Justin 15.2.1-2; Pausanias 9.7.2; Epitome, Heidelberg (FGrH 155) F1Google Scholar; Appian, , Syr. 54Google Scholar. The specific date of Alexander’s death is uncertain. Diodorus records it in 311/10, the Parian Marble in 310/09. Diodorus’ account seems to be prospective, however, recording both the peace and its consequences, and, with the peace of 311/10 probably falling rather later than generally assumed (see Wheatley, P.V., The Genesis of a King: the Early Life of Demetrius Poliorcetes, 336-306 B.C. [diss., U.W.A. 1997] 137Google Scholar), it is reasonable to postulate that, at the earliest, the young king was not disposed of until later in 310. For further discussion, see below.

14 Tarn (n.4) 23. Antigonus’ backing for the Heracles escapade is generally agreed on by scholars: e.g. Rosen, K., ‘Political Documents in Hieronymus of Cardia (320-301 B.C.)’, Acta Classica 10 (1967) 85Google Scholar; Bakhuizen (n.5) 21; Adams (n.5) 143 ff.; Brant (n.4) 32 n.l; Mendels (n.5) 175; Hammond (n.5) 164; Billows (n.l) 140; Tarn (n.4) 22; Bosworth (n.5).

15 Marble, Parian (FGrH 239) F B 18Google Scholar. The Parian Marble’s credentials for accuracy are anything but good (see, for instance, Bosworth, A.B., ‘Philip III Arrhidaeus and the Chronology of the Successors’, Chiron 22 [1992] 73–74Google Scholar; Wheatley, P.V., ‘Ptolemy Soter’s Annexation of Syria 320 B.C.’, CQ 45 [1995] 439–440CrossRefGoogle Scholar); but this entry coincides with Agathocles’ crossing to Libya which can be definitively dated by the famous solar eclipse of August 15th, 310 (Diod. 20.5.5; Justin 22.6.1), and would seem reliable. One’s confidence is shaken, however, when it is noted that the eclipse in question is recorded by this document (F B16) under the entry for 312/11. It is uncomfortably close to the archon-year change, but should at worst be placed in the year 311/10; Diodorus, probably correctly, places it in Hieromnemon’s year.

16 Diod. 20.19.1-2; 20.21; 20.27.3; cf. Justin 15.2.1-2; Orosius 3.23.36-38.

17 This seems to be indicated by the context of Diodorus: he includes it in the record for the archon year of 311/10, whereas Ptolemy’s main thrust in Pamphylia, Lycia and Caria is placed in the year 310/09. The earlier date is assumed by e.g. Kaerst, J., RE 4.2 (1901)Google Scholar, s.v. Demetrios no. 33, col. 2771; Seibert 1969 (n.5) 184-185; Hauben, H., Het vlootbevelhebberschap in de vroege Diadochentijd (323-301 voor Christus): Een prosopograflsch en institutioneel onderzoek (Brussels 1975) 52Google Scholar; Adams (n.5) 148-149; Bagnali, R.S., The Administration of the Ptolemaic Possessions Outside Egypt (Leiden 1976) 114Google Scholar. Antigonus was, by this time, in Babylonia fighting Seleucus (Babylonian Chron. 10, reverse 15-17, referring to August 310), and it can safely be assumed that Ptolemy’s timing is no coincidence.

18 Ptolemy’s son and eventual heir, Philadelphus, was born at Cos in this winter, see Parian Marble F B19; Theocr, . Idylls 17.58Google Scholar; Callim, . Hymns 4.160–195Google Scholar; Bagnali (n.17) 103. Adams (n.5) 149 n.4 also remarks on Diodorus’ sequence here, concluding that it cannot be pressed too hard to establish a relative chronography.

19 Examples in book 19 are easy to find: the archonship of Praxibulus (315/14) is introduced while Diodorus is still describing events of 316 (Diod. 19.55.1); Polemon (312/11) is introduced in the middle of Diodorus’ account of 313 (Diod. 19.77.1).

20 Wheatley (n.8). For discussion of Diodorus’ problems with chronology, see, for instance, Smith, L.C., ‘The Chronology of Books XVIII-XX of Diodorus Siculus’, AJP 82 (1961) 283–290Google Scholar; Geer, R.M.(trans.), Diodorus Siculus 9 (Loeb ed. 1947, 1984Google Scholar), Intro, x-xii; Bizière, F., Diodore de Sicile 19 (Budė ed. 1975Google Scholar), Notice x-xv; Hornblower (n.2) ch. 2; Bosworth (n. 15) 73-74.

21 So Wacholder, B.Z., ‘The Beginning of the Seleucid Era and the Chronology of the Diadochoi’, in Nourished with Peace, Studies in Hellenistic Judaism in Memory of Samuel Sandonel, ed. Greenspahn, F.E., Hilgert, E. & Mack, B.L. (Chico, Calif. 1984) 201–203Google Scholar; Hammond (n.5) 165 nn.2-3 with bibliography.

22 Diod. 19.105.2: cf. Justin 15.2.3, discussed below. The decisive fact to emerge is that Alexander’s murder was concealed, while Heracles’ was public knowledge.

23 In fact, it seems to have been a number of years before official confirmation of the news filtered through to the more distant parts of the empire. The Babylonian Chronicle of the Diadochoi (Chron. 10, rev. 34), for instance, awards Alexander IV an eighth regnal year, equivalent to April 308-March 307 (early restorations of the figure suggested the <ninth> year, see Grayson, A.K., Assyrian and Babylonian Chronicles [New York 1975] 119Google Scholar; but recent scholarship based on a joining of the two fragments of this document and consequent minimisation of the lacuna between lines 17 and 21 have led to an emendment of the restoration to <eighth> see, for instance, Geller, M.J., ‘Babylonian Astronomical Diaries and Corrections of Diodorus’, BSOAS 53 [1990] 2–3Google Scholar; Spek, R.J. van der, ‘Nippur, Sippar, and Larsa in the Hellenistic Period’, in Nippur at the Centennial, Papers read at the 35th Rencontre Assyriologique Internationale, Philadelphia 1988 [Philadelphia 1992] 249Google Scholar; Wheatley [n.13] 148-149). Consensus on the regnal years of Alexander IV has eluded scholars for over sixty years: for earlier bibliography, see Seibert 1983 (n.5) 80-81. See also Sachs, A.J. and Wiseman, D.J., ‘A Babylonian King List of the Hellenistic Period’, Iraq 16 (1954) 202–213CrossRefGoogle Scholar; Schober, L., Untersuchungen zur Geschichte Babyloniens und der Oberen Satrapien von 323-303 v. Chr. (Frankfurt 1981) 106Google Scholar ff.; Mehl, A., Seleukos Nikator und sein Reich (Louvain 1986) 129 ff.; Hammond (n.5) 167-168 n.l. Wacholder (n.21) conveniently presents the Ptolemaic and Akkadian Chronographie evidence, which almost universally postdates Alexander’s regnal years (sometimes giving him up to 11!). Mainly on the strength of this, he arrives at the conclusion (211) that Alexander IV died in 305/04. This must surely be precluded by the historical context, and would also presuppose that Heracles’ murder antedates Alexander’s; see below.Google Scholar

24 Justin’s shortcomings as a historical source have been a matter for comment over many years; see, for instance, CF. Edson’s review of Seel, O., Pompei Tragi fragmenta in CP 56 (1961) 198–203Google Scholar; Goodyear, F.D.R., ‘On the Character and Text of Justin’s Compilation of Tragus’, PACA 16 (1982) 1–24Google Scholar. However, some positive analysis has also emerged: Jal, P., ‘A propos des Histoires Philippiques: quelques remarques’, REL 65 (1987) 194–209Google Scholar. More recently, for a balanced assessment with further bibliography, see Yardley, J.C. and Heckel, W., Justin, Epitome of the Philippic History of Pompeius Trogus 1, Books 11-12 (Oxford 1997), esp. the introduction, 1–41.Google Scholar

25 Pausanias 9.7.2 blames Cassander and asserts that both of the young men were poisoned. Like Justin, he also reverses the order of the deaths, but his account is too compressed to be used for detailed analysis.

26 The promotion of Heracles must certainly presuppose the decease of Alexander IV. Heracles is frequently referred to as βασιλϵύς in the singular, and the mission as … Karáyeiv то μ€ΐράκιον em την πατρωαν βασιλϵίαν (Diod 20.20.2-3; cf. 20.28.1-2, with Macurdy, G.H., ‘Roxane and Alexander IV in Epirus’, JHS 52 [1932] 258, 261CrossRefGoogle Scholar). A living king who had been ratified since 323 and was resident in Macedonia would surely render the promotion of a dubious pretender futile. Only if the incumbent was known to be dead could the gambit be seriously enacted. For a cogent analysis, see Brunt (n.3) 31, contra Tarn (n.4). Wacholder (n.21) 204, 210, believes Alexander outlived Heracles by some years, while Hammond (n.5) 165, cf. 169 also believes Heracles died first, and places Alexander’s death between the years 310 and 306; see below.

27 Justin 15.2.3: Deinde, ne Hercules, Alexandra filius, qui annos XIV excesserat, favore paterni Hominis in regnum Macedoniae vocaretur ….

28 Diod. 20.20.1. For the circumstances and time of his birth, and an assessment of the sources, see Brunt (n.4), especially 28.

29 Plut, . Mor. 530Google ScholarLycophron, D, Alex. 800–804Google Scholar (on which see West, S., ‘Lycophron Italicised’, JHS 104 [1984] 127–151CrossRefGoogle Scholar, especially 129, 138 ff.). The public circumstances of Heracles’ demise contrast strongly with the secrecy surrounding Alexander’s. At this stage, nothing short of a public assassination would satisfy Cassander’s purposes if royalist support for the young pretender was to be stifled. Once again, this presupposes that Alexander was already dead, and there was no other candidate to whom royalists could transfer allegiance, otherwise Cassander could have promoted himself as custodian of the legitimate heir against a royal bastard pretender. The fact that he secured Heracles’ very public murder indicates he had limited his options by disposing of the other scion.

30 Hammond (n.5) 164, 169 assumes Barsine’s presence with her son, based, presumably, on the evidence of Justin.

31 Schachermeyr, F., ‘Das Ende des makedonischen Königshauses’, Klio 16 (1919) 332–337Google Scholar provides a cogent survey of the sources and arrives at this conclusion, and Tarn (n.4) 27 agrees. Justin provides a clear precedent for his confusion between Heracles and Alexander at 14.6.2, 13, which must clinch the argument. He hardly needs to be damned further by pointing out that he has already registered Polyperchon’s death some six years earlier (15.1.1). See also Hammond (n.5) 165 nl.

32 He was born within a few months of his father’s death, in autumn of 323: Curt. 10.6.9; Justin 13.2.5.

33 Diod. 20.19.2. See Wheatley (n.l3) 160-162 for discussion of the strategie tension between Polemaeus and Polyperchon. When the earlier date for the invasion is assumed, a convenient by-product is the provision of an explanation for Polemaeus’ mutiny. If Polyperchon was preparing his initiative in 310, the timing of the events is felicitous, as Polemaeus’ revolt also began in 310 in the Péloponnèse. Bakhuizen’s suggestion (n.5, 119 ff.), that Polemaeus’ action was a response to Antigonus’ negotiations with Polyperchon is worthy of consideration. On the problem of Polyperchon’s march to Tymphaea, for which we have few details, it is always possible that he proceeded via Acarnania and Aetolia, thus avoiding contact with Polemaeus.

34 Polyperchon is hardly mentioned again by the sources, which probably, but not necessarily, indicates that he played no significant part in events after 308. He appears plundering Greece with Cassander in 304/03 (Diod. 20.100.6), and was apparently still in control of parts of the Péloponnèse in 303/02 (Diod. 20.103.6-7), but disappears after this. See Heckel (n.2) 204.

35 See Tarn, , CAH 6 (1953) 493Google Scholar n.2. He also iterates this theory to Smith (n.6) 194-195, in private correspondence. The arguments are conveniently collected by Bakhuizen (n.5) 162-163; see also Bengtson (n.6) 136 n.3.

36 Diod. 20.37.1-2; Polyaenus, , Strat. 8.58Google Scholar, with Macurdy (n.9) 274-275. This must certainly have been accomplished in Polyperchon’s absence, although the status of his relationship with his son’s widow is unknown. Alexander had defected from the Antigonid alliance and his father in late 315 (Diod. 19.64.3), but in 313 Polyperchon is found living at Corinth or Sicyon, apparently in harmony with Cratesipolis (Diod. 19.74.2). The enigmatic stratagem in Polyaenus, however, seems to indicate she was glad to welcome a new partner, as does her liaison with Demetrius Poliorcetes, placed by Plutarch in 307 (Plut, . Demetr. 9.3–4Google Scholar; see below, n.41). There is also numismatic evidence for Ptolemy’s control of the region during these years: a mixed hoard of Ptolemaic and Corinthian coins from Chiliomodi, 18 kilometres from Corinth, was recovered in the early 1930’s, and it has been suggested that it comprises the savings of a soldier from the Ptolemaic garrison in Corinth; for publication and discussion of the find, see Ravel, O.E., ‘Corinthian Hoard from Chiliomodi’, in Transactions of the International Numismatic Congress, 1936, eds. Allan, J., Mattingly, H. & Robinson, E.S.G. (London 1938) 98–108Google Scholar; Mørkholm, O., ‘Cyrene and Ptolemy I, Some Numismatic Comments’, Chiron 10 (1980) 156Google Scholar; and Early Hellenistic Coinage from the Accession of Alexander to the Peace of Apamea (336-188 B.C.) (Cambridge 1991) 87.Google Scholar

37 Diog. Laert. 2.115. Plut, . Demetr. 9Google Scholar.2, however, asserts that in mid-307 Megara was garrisoned by Cassander. Ptolemy probably returned the city when he made peace with the regent before departing to Egypt (Diod.20.37.2). The order of events in Diogenes is perhaps significant: Stilpo’s negotiations with Ptolemy precede his dialogue with Demetrius, helping to anchor the excerpt to 308 and 307.

38 Suda no. 431, s.v. Δημήτριος. See Will, E., ‘Ophelias, Ptolémée, Cassandre et la Chronologie’, REA 66 (1964) 324–325CrossRefGoogle Scholar; Bakhuizen (n.5) 126-127; Bosworth (n.5).

39 Diod. 20.28.4 provides the critical, if ambiguous, Chronographie reference to the onset of winter: . This solution is suggested by Bakhuizen (n.5) 163.

40 Diod. 20.45-46; Plut, . Demetr. 8–10.Google Scholar

41 Plut, . Demetr. 9.3–4Google Scholar. The famous amorous encounter between Demetrius and Cratesipolis is barely credible in this context. The anecdote probably derives from Duris of Samos: see Sweet, W.E., ‘Sources of Plutarch’s Demetrius’, CW 44 (1951) 177–181Google Scholar; Hornblower (n.2) 225-226. Demetrius is generally depicted as focussing single-mindedly on either military or frivolous matters, but not combining the two (Diod. 20.92.4; Plut, . Demetr. 19Google Scholar.3), and I would suggest that the incident has been lifted from its correct historical context by Plutarch in order to begin his documentation of the perversity of Demetrius’ character; for discussion, see Wheatley (n.13) 174-175.

42 The reality, of course, was that Demetrius arrived in Greece with an overwhelming armada, and received the immediate submission and co-operation of the Athenians, thus firmly establishing a new Antigonid presence in the region which was to remain until the battle of Ipsus in 301.

43 Polemaeus was evidently present in Iasus either just before being summoned to Cos, or in Ptolemy’s service after arriving, as his name appears on an inscription as one of the negotiators of a treaty between that city and the Lagid: IGSK 28.1 no. 2. See Hauben, H., ‘On the Ptolemaic lasos inscription IGSK 28.1, 2-3’, EA 10 (1987) 3–5Google Scholar, with further bibliography, n.l; and Bosworth (n.5).

44 I am indebted to Professor A.B. Bosworth for reading and commenting on an earlier draft of this paper; also to Associate Professor G.R. Stanton and the anonymous reader for Antichthon, whose suggestions markedly improved the original. The remaining infelicities are, of course, my own responsibility.