¹ For evidence that three victories in the pentathlon constituted overall victory see Pollux, Onomasticon iii 151, … ἐπὶ δὲ πεντάθλου τὸ νικῆσαι ἀποτριάξαι; scholion to Aristidcs, Panathenaicus 339, … ἀρκεῖ γὰρ αὐτοῖς γ’ τῶν ε’ πρὸς νίκην. See also Pausanias iii 11.6 where Hieronymos of Andros defeats Tisamenos of Elis 3–2 and Bacchylides 9 where Automedes of Phlious emerges victorious by winning in the two throwing events and in wrestling.

² Philostratos, Gymnastikos 3, in a passage to be discussed at length below, is our best witness for this fact. See also Merkelbach, R., ‘Der Sieg im Pentathlon’, ZPE xi (1973) 261Google Scholar, for several ancient references to the second-class abilities of pentathletes.

³ For a good summary of scholarship in the two schools of thought regarding victory in the pentathlon see Bean, G. E., ‘Victory in the pentathlon’, AJA lx (1956) 361–8Google Scholar. After Bean's study, Harris, H. A. published Greek athletes and athletics (London 1964Google Scholar). On pages 77–80 he suggested that only victors in the first four events competed in wrestling, others being eliminated.

This theory was accepted by Merkelbach (see n.2). In his Sport in Greece and Rome (Ithaca 1972) 34–35, Harris re-evaluated his previous stand and offered the more attractive theory that only winners of the first three events went to the race and wrestling competition.

⁴ The early history of the theory of relative finish and its subsequent complication by the addition of numerical values can be found in Bean's article cited above. Since Bean's study, Ebert, J., ‘Zum Pentathlon der Antike’, Abhandlungen saechs. Akademie der Wiss. zu Leipzig, phil.-hist. Klasse, Band 56, Heft 1 (1963)Google Scholar, has suggested that a pentathlete was eliminated whenever he was beaten three times by any other competitor. This theory is what prompted Merkelbach's article, cited above. Ebert answered Merkelbach's objections in ZPE xiii (1974) 257–62. A new twist to this theory has been offered by Sweet, W., Sport and recreation in ancient Greece (Oxford 1987) 56–9Google Scholar. Rather than keep count of second place finishes, Sweet suggests a repechage of early events, now lacking the former winners. For objections to various aspects of Sweet's theory see Langdon, M. K., ‘Scoring the ancient pentathlon: Final solution?’ ZPE lxxviii (1989) 117–118Google Scholar.

⁵ Followers of the relative finish theory have historically placed great faith in Philostratos' testimony and have therefore had a high regard for second place finishes. Progressive elimination theorists, on the other hand, are convinced that only first place finishes were significant and have consequently had little regard for Philostratos. Philostratos is certainly not beyond reproach. For a good resumé of faults in his treatise see Poliakoff, M., Studies in the terminology of the Greek combat sports (Koenigstein 1982) 143–8Google Scholar.

⁶ The rigors of mastering five different skills could not have encouraged large numbers of athletes to become pentathletes. Harris also tells us (Sport in Greece and Rome, 34) that prize money for the pentathlon was only a quarter ofthat offered for the combat sports at the beginning of the present era. Faber, M., Philologus 5 (1891) 492fGoogle Scholar, and Gardiner, N., JHS xxiii (1903) 61Google Scholar, insisted that the pentathlon probably seldom featured more than a dozen participants.

⁷ It is impossible to say whether Philostratos believed that the origin of the pentathlon lay in the myth which follows—nor is his belief important to us. He thought the story worth telling. No great honor attaches to the pentathlon from having been invented by Jason or having been won by Peleus. Philostratos' reason for telling the story must be that it accounts for the glorification of the second-class athlete whom the competition would attract and that it contains within itself a way by which the best among seconds could achieve victory.

⁸ Herein lies the basis for the theory of relative finishes and the antipathy of progressive eliminationists towards Philostratos.

⁹ The supposition required here is not gratuitous. Clearly Philostratos expects his contemporaries to be familiar with the basic rules by which the pentathlon was contested. This accounts for the bare presentation of the myth. It is we who are in the dark and have been confounded by his, for us, unclear presentation. The point of this paper is that Jason would have mandated early elimination as a feature of this new contest, a feature well-known to Philostratos and his contemporaries. Immunity from elimination was, of course, granted to a winner of any event until one man had accumulated three victories. This point is agreed by both schools of thought. Thus, if the winner of the first event finished last in the second, he nonetheless was included in the number of those going on to event number three, the man finishing last but one being eliminated.

¹⁰ Ebert, Merkelbach and other investigators of the pentathlon have been strongly influenced by the contention of Juethner, J., Philostratos ueber Gymnastik (Leipzig 1909) 193Google Scholar, that δεύτερος in the passage cited above was used by Philostratos to mean ‘inferior’, not ‘second best’. To support this contention Juethner cites five occurrences of … δεύτερος in the collected works of various Philostratoi as they appear in Kayser's edition of 1840. Two occur in the Life of Apollonius of Tyana and clearly mean ‘second’. The other three are neuter plural in form and carry the significance Juethner suggests. We should not, however, try to force this significance on singular forms. I. Avotins and Avotins, M. M., An index to the Lives of the Sophists of Philostratos (Hildesheim 1978) 68Google Scholar, list nine occurrences of δεύτερος in this work in addition to the two cited by Juethner. All nine are singular and mean ‘second’. Juethner's own Wortregister for Gymnastikos lists two other places where our Philostratos used δεύτερος, both meaning ‘second’. In addition, had Philostratos wanted to imply a general inferiority of Peleus, he would have used δεύτερα as a pendant to the earlier κράτιστα or a suitable antonym, χείρονα or ἤσσονα, not the singular δεύτερος which can here, as elsewhere, mean only ‘second’.

The interpretation offered here obviously depends upon the five contestants listed by Philostratos being the only contestants, a fact, as far as I know, acceptable to all. The sense of δεύτερος demonstrates that Philostratos does not mean that the sons of Boreas were both best at both running and jumping. Also, in no case could ties or dead heats be allowed in the pentathlon. See Harris, H. A., JHS lxxxii (1962) 21–4Google Scholar.

¹¹ Practical application appears to have played a small role in the development of some earlier theories. We should expect the pentathlon to be carried out expeditiously and in a way which would be readily appreciable to even casual spectators. For the need to complete the pentathlon in a reasonable length of time see Pausanias v 9.3 and the forced rearrangement of events when the pentathlon intruded upon the running of more popular events. In addition, most of the theories of relative finish are too complicated to be easily employed by any but the most dedicated fans of the sport.

¹² It was the realization that those who specialized in the sprints and those who regularly competed only in wrestling would have an unfair advantage if elimination was held off until the completion of the stade race which led Harris, , in Greece and Rome xix (1972) 64Google Scholar, to revise his earlier theory. An especially skilful sprinter or wrestler might well try to invade the pentathlon if safeguards were not built into the competition.

¹³ For the use of markers in competition see Homer, Od. viii 192ff. where the discus cast of Odysseus is marked by Athene. See also Swaddling, J., The ancient Olympic Games (London 1980) 55Google Scholar, for a photograph of a jumper with markers in a vase painting. It might be useful to distinguish winners of the earliest events from those subject to elimination. This could be accomplished by a fillet tied around the thigh of the jump winner or the non-throwing arm of the discus and javelin winners. Vase paintings of filleted athletes are common, but they seem never to have been studied from an athletic viewpoint. See Juethner, J., ‘Siegerkranz und Siegesbinde’, Oesterreich. Archaeolog. Institut. Jahreshefte i (1898) 42–48Google Scholar.

¹⁴ Whether a total of four competitors or the four best discus throwers and the best jumper (if he did not finish in the top half of the discus field) passed to event number three is not especially troublesome. It would not be surprising if practice varied from place to place and time to time.

¹⁵ The nature of the triad leads one to believe that most commonly three men survived to take part in the fourth round. Jumping called for good legs. The throwing events required a strong arm, but two very different techniques. We know from ancient sources that sometimes one man triumphed ἐν τῇ πρώτῃ τριάδι, but the hope of games organizers and pentathlon fans must have been a full five event competition. Thus, even if a strong-armed thrower won events two and three, the stade race would favor a jumper and lead to the tie-breaking wrestling final.

¹⁶ Dissatisfaction with what was felt to be excessive importance placed upon wrestling skill goes back to Percy Gardner in 1880 and has continued, apparently being Sweet's motivation for backing a replay of events from the triad. Merkelbach, 265, maintains that a double winner entering the wrestling against two single winners could do no worse than tie for the championship. His belief that a two-time winner had to wrestle twice (264 n.8) is based on doubtful evidence, a lacunose line in which his key word σώματα is the suggested reading for γυια[…]ματα. Harris sees no problem in the setup outlined here, nor does Langdon, p.118.

¹⁷ An anonymous referee to whom I am exceedingly grateful for numerous suggestions concerning this paper has posed an interesting question. We are told in some ancient inscriptions that prizes were sometimes offered to second place finishers in the pentathlon. How were they determined? I believe the most likely candidate for this honor would be the loser of the final wrestling match. He would always have two wins, while the overall winner would have three. If the pentathlon ended after the stade race, the runner-up would be the winner of the one event of four not won by the overall victor.

¹⁸ Jason's less than heroic nature in his dealings with Medea is well known from the tragedies of Euripides and Seneca. For a good description of Jason in Hellenistic epic see Lawall, G., ‘Apollonius' Argonautica: Jason as anti-hero’, YCS xix (1966) 121–09Google Scholar. Interesting from our point of view is Lawall's observation that Apollonius has little use for the specialized skills of the Argonauts and that Jason exploits love (136). He notes also that Lemnos is a place where the resourceful Jason begins to emerge (151) and that it is a mark of Jason's character that he sacrifices heroic values for success.