BMCR 2004.03.09

Boats of the World: From the Stone Age to Medieval Times

, Boats of the world : from the Stone Age to Medieval times. Oxford: Oxford University Press, 2001. 1 online resource (505 pages) : illustrations, maps. ISBN 9780191590535. £120.00.

I offer my apologies to all readers for the tardiness of this review.

In this ambitious book on boat archaeology Seán McGrail (henceforth McG.) succeeds at covering many centuries of shipbuilding (and seafaring) history, basically looking at developments all over the world, excluding the coasts and rivers of the former USSR and the African continent except for Egypt but providing detailed diachronic studies of Egypt (pp. 14-54), Arabia (pp. 55-87), the Mediterranean (pp. 88-165), Atlantic Europe (pp. 166-248), India (pp. 249-278), Greater Australia (pp. 279-288), South-East Asia (pp. 289-310), Oceania (pp. 311-345), China (pp. 346-393), and the Americas (pp. 394-430). These chapters are framed by insightful sections on sources and themes (pp. 1-13) and early water transport (pp. 431-439). The book is rounded off by a rich bibliography (pp. 441-465), a glossary (pp. 466-470; particularly useful for habitual landlubbers like the present reviewer), and a carefully compiled index (pp. 471-480). Indeed, McG. covers all conceivable kinds of watercraft, ranging from Indian pot boats (p. 266) to Vietnamese basket boats (pp. 294-296) and European cogs (pp. 232-239).

The arrangement by geographical regions, though not the only one possible, is perhaps the most efficient in a book like this. Inevitably, this practice leads to some repetitions, for example when McG. uses the Periplus of the Erythraean Sea as a source text in his chapters on Egypt (pp. 51-53), Arabia (pp. 77-80), India (pp. 256-260; in particular pp. 256-257 partly duplicate what has already been said on p. 51, and some of the general information on the nature of the periplus literature should have been transferred there), South-East Asia (pp. 291-292), and China (p. 387). Apart from such trivia, the work is a success: McG. has to be congratulated on his impressive achievement. Yet if you want to buy this weighty (and huge) monograph — and anyone interested in the archaeology and cultural anthropology of shipbuilding and seafaring should consider its purchase — make sure you have an appropriately sized shelf at home.

This highly informative book is instructively and lavishly illustrated. It contains judiciously chosen and well-presented facts and theories and is excellently written — in that rare kind of style which is at the same time precise, lucid, simple (without being simplistic) and clear — a pleasure to read. However, the specialist in maritime archaeology will certainly benefit more from reading it than a complete novice in that field. In particular, experts will be able to appreciate more fully some of the finer technical details. Being a classicist, my primary concern is with Greek and Roman antiquity, and as someone who is currently working on maritime and nautical terminology and imagery in Roman poetry I am particularly interested in what maritime archaeology can tell me about the Realien of ships, shipbuilding, and seafaring to which the literary sources refer.

Since I have nothing serious to disagree about with McG., I will talk only about his use of sources and mention a few points where one might profitably expand (or clarify) his argument.

In Chapter 1: “Sources and Themes”, McG. claims (on p. 12) that it is “[t]he aim of this study to use all forms of evidence, especially archaeological, to present an account of how rafts, boats, and ships were built, propelled, steered and generally used, from earliest times to somewhere in the period AD 1400-1800.” Evidence allowing, he also discusses “methods of navigation, means of exploration, and principal overseas trading routes”. On pp. 1-5 McG. specifies the sources of evidence available to us. The primary source is, as a matter of course, material won by archaeological research (especially “excavated remains of water transport”, p. 1); other sources are indirect evidence (e.g. “distribution patterns of artefacts”, p. 2) for early overseas voyages (and thus for the existence of seagoing vessels), iconography (a very good example of how such evidence may be used is given by McG. on p. 233), (written) documents (such “evidence ranges from inscriptions mentioning shipping, and early law codes listing harbour dues, to detailed technical reports written and illustrated by explorers and travellers”, p. 3), ethnography (mainly the documentation of vessels “still in use in non-industrial, generally illiterate, small-scale societies”, p. 3), scientific dating (radiocarbon assay, dendrochronology), environmental evidence (form of the coastline, river gradients, existence of reefs, shoals, sands etc.).

Except for evidence that is won by scientific dating and the study of past environmental conditions, McG. remains sceptical of other than archaeological sources. He is particularly suspicious of iconographic and documentary evidence (“cannot be accepted without rigorous analysis and interpretation”, p. 2; “unlikely to be comprehensive or unbiased”, p. 3) and indirect evidence for early seafaring (“must be examined critically and, in the case of non-islands, the balance of probability struck”, p. 2). This may explain why McG. does not readily accept experimental voyages as they have been undertaken by Thor Heyerdahl or Tim Severin as evidence for the scope of early seafaring: such work “tells us little about early maritime history, but does tell us something about the seagoing capabilities of log rafts and the skills of twentieth-century human beings” (p. 387; similarly sceptical views, especially on Heyerdahl, can be found on p. 397f.). I feel inclined to be significantly less sceptical (just think of the success of the Olympias project — the full reconstruction of a Greek trireme in spite of the absence of any direct evidence; see pp. 141-144); still, McG. argues his point convincingly.

McG. mentions boat models as a possible source of evidence, but only briefly. While it is quite possible that most watercraft models do not provide us with sufficient (or sufficiently reliable) information about the actual properties of the vessels they are meant to represent, the possibility that some of them betray certain features which might in turn lead to new insights into their architecture should be seriously considered (a good example from Egypt is provided by McG. himself on p. 159).1 The earliest example of Mesopotamian watercraft is a clay model.2 With all due caution, the role of boat models as a source of interesting supplementary evidence (for example for the housings of stern oars or the length to beam ratio) should not be underrated. It is of course understandable that one might feel reluctant to refer to boat models when dealing with periods for which ample direct evidence has been won by the excavation of watercraft remains, but one can certainly use these models for periods where such (‘hard’) evidence is lacking.

Although McG. does not explicitly mention (fictional or non-fictional) literature, he shows himself generally well-informed about the relevant literary sources (see, for example, p. 126 on Homer’s ships). Sometimes he could make more of them, but, as a rule, his treatment of them is reliable.

Now for some details.

p. 47f.: It cannot be known for sure whether the story of the Egyptian official Wenamun, sent on a business mission to the Levant by the High Priest Herihar (probably about 1076/1075 BC is literary fiction wholesale or not. Still, the narrative appears to reflect the actual political and economic climate of the time quite well, and a voyage like that of Wenamun is at least perfectly conceivable.

p. 98: There is a more recent study than Henkel (1900) on the limits of visibility from sea level in the Mediterranean which McG. might have mentioned: G. Schüle, Navegación primitiva y visibilidad de la tierra en el Mediterráneo, in: Crónica del XI Congreso nacional de Arqueología, Zaragoza 1970, pp. 449-462.3

p. 100f.: Discussing early navigational techniques, McG. states that “[t]hroughout the world, as demonstrated in other chapters of this book, from the earliest times that there is evidence until well into the medieval period, seamen used non-instrumental navigational techniques, based on inherited traditions, personal experience, and detailed observation of natural phenomena.” And he adds: “The only seagoing navigational aid known anywhere in the world before the ninth century AD is the sounding lead and its near relative, the sounding pole.” By and large, this is certainly true. Still, the recovery of a complicated mechanical device of the first century BC, which is usually regarded as an astronomical clock, may be taken as an indication that we do not know the whole history of Greek mechanics yet.4

p. 126 (and elsewhere). For a reader who does not have Rieu’s (outdated) Penguin translations of the Iliad and the Odyssey on his desk, it is just impossible to trace references given to the page numbers of these translations. It would have been much better to quote — as is standard practice in Classics — by book and line number, as McG. himself also does sometimes (for example, in the second paragraph).

p. 126 (on Hom., Il. 2.135). McG. has put his finger on a very difficult problem, namely the meaning of the Homeric hapax σπάρτον. From the context alone it is impossible to decide whether the word refers to the rigging (as many have thought) or to the cords (as McG. takes the word). At any rate, two authorities of such repute as Varro (ap. Gell. 17.3.4) and Pliny, Nat. hist. 24.65 take the σπάρτα of Il. 2.135 indeed to be cords used to sew together the ships’ planking. In addition, there appears to be fresh evidence that “sewing as a means of fastening planks to each other continued throughout ancient times for certain types of craft”, see Lionel Casson, Ships and Seamanship in the Ancient World, Baltimore and London 1995 (originally published 1971; new ed. with addenda), p. 443. Still, the case is adhuc sub iudice, see Ch. Kurt, Seemännische Fachausdrücke bei Homer: unter Berücksichtigung Hesiods und der Lyriker bis Bakchylides, Göttingen 1979, p. 163f.

p. 133. On the terminological difficulties in Ezekiel 27, the only description in toto of a ship in the Old Testament, see also the perceptive study of E. Strömberg Krantz, Des Schiffes Weg mitten im Meer. Beiträge zur Erforschung der nautischen Terminologie des Alten Testaments, Lund 1982, pp. 75-151.

p. 136-138. One of the great assets of this book is that McG. provides very often valuable cross-cultural overviews like this one (on the use and construction of sewn-plank boats in Roman and Mediaeval times).

p. 138 (Odysseus’ boat; Hom., Od. 5.244-257). McG. informs his readers about the scholarly debate concerning the actual way in which Odysseus went about constructing his boat. Casson (see above, n. χχχ pp. 217-219 and 461, is probably right in taking the passage (to which McG. variously refers as “234-257” on p. 126 and “244-257” on p. 138) to mean that Odysseus should be envisaged as hammering the craft with pegs and joints, and not as sewing it. Still, meaning and syntax of some of the words are quite obscure. At any rate, the Odyssey is not meant to be a shipwright’s manual, and so I have serious doubts that any attempt at explaining what is ‘really’ going on when Odysseus puts together the various parts of his boat will ever be completely satisfactory. To my view, the most important thing about this passage is that through detailed and controlled narrative the poet conjures up bit by bit an impressive picture of how a well-made and reliable boat is — somehow, for who would know the difference? — assembled, a craft quite fit to carry Odysseus away from Ogygia and back home. This, in turn, makes for a very effective contrast when, shortly afterwards, the audience are told how Poseidon destroys that same watercraft (Od. 5.282-332) — again, bit by bit. Discussing the various interpretations given to this passage, P. Janni, Carpenteria navale e scrittori antichi, in: I. Mazzini (ed), Civiltà materiale e letteratura nel mondo antico. Atti del seminario di studio, Macerata, 28-29 giugno 1991, Macerata 1992, pp. 45-53, notes rightly (on p. 48) that it is “il solito errore di oggettivizzare troppo ciò che un poeta descrive.”

p. 221: “Syldt” should be spelt “Sylt”. An interesting addition to the information given about the Frisians as skilled seafarers is that in the course of the eighth century, the notion of skilful seamanship became almost synonymous with ‘Frisian’. This reputation was to last until long after other nations dominated the seaways, although the Frisians probably never had much sea power in their own right,.5

p. 233. A detailed study of the 1329 town seal of Stralsund is provided by D. Ellmers, Die mittelalterlichen Stadtsiegel mit Schiffsdarstellung an der südlichen Ostseeküste, in: H. bei der Wieden (ed), Schiffe und Seefahrt in der südlichen Ostsee, Cologne and Vienna 1986, pp. 41-88 (the 1329 Stralsund town seal is treated on pp. 69-71). Ellmers shows convincingly that the depiction of ships on seals is not only influenced by the artist’s effort to portray the ship realistically but is also (as just as much) influenced by iconographic conventions, artistic traditions, and pictorial symbolism (p. 42). In other words, these seals do not necessarily depict a particular existing ship of the real world, but rather an ideal ship that was actually made up of elements (primarily?) taken from other depictions of vessels of the same type on seals, probably combined with some innovative elements, introduced for the sake of artistic competition and with the intention of the government of one city to outdo another (Ellmers says that the Stralsund seal was, for example, modelled on the 1281 seal of Lübeck). These innovative elements may in turn be derived from (and mirror developments in) the real world and betray signs of certain innovations in the naval architecture of their times (as is indeed the case with the Stralsund seal). This does, of course, not diminish the value of such iconographic evidence at all. Rather, one should bear in mind — and McG. is largely conscious of such things (see his statements on p. 2f.) — that in the first instance they provide evidence for what certain people at a given time thought possible (and conceivable) in the field of ship-building rather than for a particular individual ship.

p. 256: The position of the harbour city of Myos (or Aphrodites) Hormos on the map of the Indian Ocean is puzzling — it should be in Egypt (on the West coast of the Red Sea, north of Berenike), not on the Arabian coast. The correct position is indicated on the map of Egypt (on p. 15). Was it here mixed up with Leuke Kome?

p. 256f.: McG. calls the Periplus Maris Erythraei“an example of a group of early Mediterranean texts called periploi or ‘circumnavigations’ which gave information about harbours and watering places along a particular coast” etc. Actually, the Periplus of the Erythraean Sea itself is a rather late example of the genre, showing a revival, probably induced by the growing demand for such books due to the expansion of Roman overseas trade in Asia. The first literary version of a periplus that is known to us goes under the name of Scylax of Caryanda (roughly mid-fourth century BC; mentioned by McG. on p. 160), see OCD (3rd ed) s.v. ” periploi“. The technique of reporting one’s ‘voyage around’ is certainly much older, probably going back to Phoenician times. Traces of it are also found in Homer, see K. E. Müller, Geschichte der antiken Ethnologie, Reinbeck 1997, pp. 43f., 54f., 70f.

p. 438 (cf. also p. 101, where McG. rightly states that Homer “has a ‘mental chart’ in his head, giving him the spatial relationships of the coastal lands and the islands of the Mediterranean”): In the context of the question how navigators might have constructed a ‘mental chart’, mention could have been made of some of the more recent literature dealing with that topic, for example a study by A. Wolf, who has tried to show that Iliad and Odyssey were written by someone who “fulfilled the principal precondition of map-drawing, that is: he possessed the ability to picture the individual shape of a land or its coast-line from a bird’s eye view, i. e. to imagine a map.”6 It would appear that there is still a lot of research to be done in this promising field.

pp. 441-465 (bibliography): Considering that this is a work that is intended to study how rafts, boats, and ships were used in many cultures and regions at different times, it is not to be expected that the bibliography will be exhaustive, and in fact it should not be. In general, the bibliographical items provided by McG. are extremely judiciously chosen. Apart from some of the books mentioned in the course of this review, there are only very few titles that I think might profitably be added to his list:

J. H. D’Arms, E. C. Kopff (eds), The Seaborne Commerce of Ancient Rome: Studies in Archaeology and History, Rome 1980.

E. Dodd, Polynesian Seafaring, New York 1972.

R. Feinberg, Polynesian Seafaring and Navigation, Kent 1988.

O. Höckmann, Antike Seefahrt, Munich 1985.

I. Pekáry, Repertorium der hellenistischen und römischen Schiffsdarstellungen, Münster 1999.

J. Rougé, La marine dans l’Antiquité, Paris 1975.

J. Rostropowicz (ed), Morze w kulturze starozytnych greków i rzymian, Opole 1995.

pp. 471-480 (index): The usability of the book is somewhat marred by the fact that there appears to be no separate index of the numerous illustrations, nor of the literary passages referred to.

“Of the unique importance of water transport to mankind there can thus be no doubt. But this importance is not reflected in the knowledge we have today of early maritime matters.” (McG. in the final paragraph of his work on p. 439) With this admirable book, McG. has contributed a lot to the dissemination of such knowledge as well as to the stimulation of new research. He has completed his arduous task most successfully, displaying his proven scholarly expertise in the field of maritime archaeology as well as conveying his great enthusiasm for it.


1. See P. F. Johnston, Ship and Boat Models in Ancient Greece, Annapolis, Maryland 1985, a work that is unfortunately not mentioned by McG. For an instructive clay model of a heavy merchant galley from the 9th/8th century BC, see L. Casson, Ships and Seamanship in the Ancient World, Baltimore and London 1995 (originally published 1971; new ed. with addenda), p. 65f.; on seventh century wooden models from Samos, see Casson, op. cit., p. 55 n. 72.

2. Casson, op. cit., p. 22.

3. A thorough monograph dealing with all these questions is now J. Morton, The Role of the Physical Environment in Ancient Greek Seafaring, Leiden/Boston/Cologne 2001.

4. On this remarkable mechanical device see the detailed study by D. de Solla Price, Gears from the Greeks: The Antikythera Mechanism — A Calendar Computer from ca. 80 B.C., Philadelphia 1974 ( Transactions of the American Philosophical Society, New Series, vol. 64, part 7). Although it is unlikely that this device was used for purposes of facilitating navigation (see Price, p. 22), the possibility cannot be ruled out entirely that mechanisms of a simpler kind existed and were indeed used for that purpose.

5. See A.-E. Christensen, Vikingtidens Denmark, Copenhagen 1969, pp. 146-164.

6. A. Wolf, Hatte Homer eine Karte? Beobachtungen über die Anfänge der europäischen Kartographie, Karlsruhe 1997, p. 66. One may be reasonably sceptical about some (certainly not all) of Wolf’s findings, but even if nothing else, his considerations on mental charts definitely deserve close attention.