How the Greeks Sailed into the Black Sea

Author(s): Benjamin W. Labaree

Source: American Journal of Archaeology , Jan., 1957, Vol. 61, No. 1 (Jan., 1957), pp.
29-33
Published by: Archaeological Institute of America

Stable URL: http://www.jstor.com/stable/501077

REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.com/stable/501077?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and

extend access to American Journal of Archaeology

This content downloaded from

62.29.13.150 on Thu, 27 Aug 2020 13:08:29 UTC
All use subject to https://about.jstor.org/terms
 How the Greeks Sailed into the Black Sea
BENJAMIN W. LABAREE

Oth. .... Like to the Pontic sea, with their own colonies and with other peoples set-
Whose icy current and compulsive course tled along the shores of the Black Sea.4 Thucydides
Ne'er feels retiring ebb, but keeps due on
traced these developments,5 and Isocrates among
To the Propontic and the Hellespont. ....
others indicated that the presence of Greek trading
IN THE leading article of the Symposium on Homer
vessels in the Black Sea was by no means unusual.6
(AJA 52 [1948] i-io), Rhys Carpenter raisesIfthe the Bosporus was impassable to the merchants'
question of how Greek shipping penetrated the vessels, how was this trade carried on?
sailing
Black Sea.2 He cites three basic factors which Carpenter
make goes on to say that it was only with
this voyage difficult: i) during the sailing months
the development of the first powerfully-oared ves-
of the late spring and summer, winds in the sel,Bos-
the pentekonter, that the Greeks possessed a
porus prevail from the northeast, that is, from the
ship of sufficient speed to overcome the Bosporean
current. He demonstrates how it must have been
Black Sea into the Sea of Marmara; 2) a formidable
current runs through the Bosporus into thethe Seapentekonter,
of not the trireme, which was the
Marmara, representing the excess water from invention
the of Ameinokles in about 68o B.c. This is
several large rivers which flow into the Black Sea,
a welcome new suggestion. With a vessel at their
in particular, the Dnieper, Bug, Dniester, and
disposal which could be propelled independently
Danube rivers: this current is at its strongest,of the winds, Greek merchants no longer found the
aver-
aging four knots, during the sailing season,Bosporus
partly an impassable barrier. Carpenter con-
because the northeast winds help to pile upcludeswaterthat only with this powerfully-oared vessel
at the Bosporus end of the Black Sea, further in-the region beyond the Bosporus be reached
could
creasing its level in relation to the Sea of Marmara;
by water, thus explaining why there is no evidence
3) the ancient sailing vessel, rigged withofbut colonization
a in the Black Sea area before about
680 B.c.
single square sail, could not beat against the north-
east winds, nor was its rowing speed nearly great
The difficulty is that, according to the thesis,
enough to overcome this four-knot current.not
Hence
only the first penetration of the Black Sea, but
likewise all subsequent passages through the Bos-
"... it was never possible to sail through the straits.
porus could have been negotiated only by a power-
Yet it is well known that from the seventh cen-
fully-oared vessel such as the pentekonter or the
trireme. This assumption must be made because
tury on the Greeks established and regularly traded
1 Othello III, iii. trade often preceded actual colonial settlement. For all this
2 The present author is deeply grateful to Professor Sterling
see T. J. Dunbabin, The Western Greeks (Oxford 1948). For
Dow, of Harvard University, for suggesting this reexamination
Greek colonization of the Black Sea shores E. H. Minns, Scy-
of the problem raised in Carpenter's article and for assisting
thians and Greeks (Cambridge, Eng. 1913); M. Rostovtzeff,
him with references to the several secondary works cited.
Iranians and Greeks in South Russia (Oxford 1922). On the
8 Rhys Carpenter, "Greek Penetration of the Black Sea," AJAand varied fifth- and fourth-centuries trade to and
extensive
52 (1948) I. from South Russia, see M. Rostovtzeff, Social and Economic
41It need only be mentioned here that the openingHistory
of theof the Hellenistic World (Oxford 1926) 105-111 and
Black Sea was frequently reflected in mythology, most con-
notes. For the south shore, especially what was later the King-
spicuously in the story of the Argonauts and the wanderings
dom of Pontos, see D. Magie, Roman Rule in Asia Minor
of Odysseus before western adventures were added.(Princeton
An ex- 1950) 182-186. For summary see CAH III, 657-
cellent summary of these myths is contained in H. J. Rose,
668. To the bibliography (755-756) much is to be added. For
A Handbook of Greek Mythology (New York 1929)instance,196ff it has been shown that in the Tribute Assessment
for the Argonauts and 243ff for Odysseus. Decree of 425/4 the cities assessed in the Euxine panel re-
In the present study, it is not my purpose to attempt
quiredthe
not forty-four lines (meaning very nearly forty-four
establishment of a date for the earliest penetration of the Black
cities) but rather from forty-eight to fifty-one lines. S. Dow,
Sea but merely to show that from the first development TAPAof the
72 (1941) 83-84.
sailing vessel such penetration was possible. The actual dateI3-14.
5 Lines
on which this was first accomplished can only be established
I Isocrates, Trapeziticus 57. See also Aristotle, Economicus
on the basis of archaeological evidence from Black Sea settle-
2.1I; Herodotus 4.24, and 7.I47.
ments. In fact, recent study of western colonies has shown that

This content downloaded from

62.29.13.150 on Thu, 27 Aug 2020 13:08:29 UTC
All use subject to https://about.jstor.org/terms
30 BENJAMIN W. LABAREE [AlA 61

0.0-BLACK S

as 94

HE

,No

NW~ONDI~CH

This content downloaded from

62.29.13.150 on Thu, 27 Aug 2020 13:08:29 UTC
All use subject to https://about.jstor.org/terms
1957] HOW THE GREEKS SAILED INTO THE BLACK SEA 31
Not affecting
there were no fundamental changes only did the presence
the of rowers eliminate
basic performance of sailing vessels
thethroughout the
possibility of combining swift propulsion with
ancient period. In short, Carpenter's argument
cargo-carrying, but theis:cost of operating a rowed
throughout antiquity, no vessel could sail into the
vessel was prohibitive to all but a very few wealthy
Black Sea. individuals. The monthly payroll for a trireme in
Other readers must have felt the objection to commission has been estimated by Thucydides at
this conclusion which prompted the present study.the rate of a drachma a day for each sailor."3 On
Pentekonters and other rowed vessels were devel- this basis it would have cost well over one talent
oped for a specific purpose: to provide a high-speed to maintain a pentekonter during the sailing season,
vessel, not dependent upon the winds, with whichan extremely high fixed cost for any merchant. The
to gain control of the seas by force. The very de-very factors which gave the long ship its admirable
sign of these ships departed radically from the de- fighting qualities rendered it unfit for use as a
sign of the typical merchantman. Where the latter commerce carrier.
was broad of beam and deep of hold to provide If the long ships were unsuited for trade and
maximum cargo space, the warship was long, nar-the round ships incapable of sailing to windward,
row, and of shallow draught to permit high-speedespecially against a strong current, how then did
propulsion. The distinction reveals itself in manythe Greeks carry on commercial relations with the
ways, perhaps most clearly in the names given byPontus? In order to answer this question, we must
the Greeks to each type of vessel. The merchantclosely examine conditions in the Bosporus.
ship was generally referred to as the round ship,' The current which renders these straits difficult
the largest being known as the tubs.8 War vessels,of passage is caused principally by two factors."
on the other hand, were called long ships,9 withThe several rivers of Eastern Europe which flow
various types given such names as the shark?0 or into the Black Sea carry their largest amount of
the racehorses.n" waters when draining off the melted snows of their
The reason for this distinction has always been watersheds. This excess brings the Black Sea to its
clear. Speed and maneuverability were priority re-maximum level from the latter part of June to the
quirements for the warship, whereas economy ofend of August.' The second, and much more im-
operation was of secondary importance. Sea-borneportant, element is the action of winds blowing
trade, on the contrary, demanded a merchant ves-from the northeast during this season. These winds
sel of large capacity and low-cost operation. Neither tend to push the surface water of the Black Sea
of these prerequisites could be obtained by the usein a southwesterly direction-toward the Bosporus
of the long ship, since the very factors making for -while at the same time the Sea of Marmara is
swiftness meant a reduction in capacity and an in-similarly concentrated at its Hellespont end. A con-
crease in cost of maintenance. The long, narrow siderable current is therefore produced as Black
shape of its hull provided a minimum of space, Sea waters rush through the narrows to seek a
almost entirely occupied by the many rowerslower level in the Sea of Marmara. The interaction
needed to attain high speed. So crowded were con-of these two factors can be simply stated: while
ditions on board a trireme that few provisions forthe excess of Black Sea water gives the Bosporus
the crew could be carried along. When such a its "average" current of about three knots,"' it is
vessel was used as a cavalry transport, only sixty primarily the northeast winds which accelerate its
oars could be operated, since the cargo of thirtyvelocity to the five or six knots sometimes observed
horses entirely filled the hold.'2 A rowing vessel,in parts of the channel.7
then, when converted to a cargo carrier, could no Since the major rivers flow into the northern half
longer enjoy its original speed. of the Black Sea, their waters will be affected by
7 C. Torr, Ancient Ships (Cambridge 1894) 23, footnote 59. Pilot relative to currents in the Bosporus a
(arporfb16x). of observations made by the British naval ve
s ibid. 113. (yav"Xot). the periods August 17 to 26 and October 1
9 ibid. 23, footnote 59. (~caKpd). fact that these are not "up-to-date" findin
lOibid. 121. ( rplareLt) I ibid. io8. (KeX'r7a). bearing on the applicability of these conclus
1 ibid. 14. 13 6.31. of Bosporean currents in ancient times.
14 U.S. Hydrographic Office, Black a5
Sea ibid. 10 ibid.
Pilot (Washington
1926) 45. References to conclusions17 made
ibid. 49. in the Black Sea

This content downloaded from

62.29.13.150 on Thu, 27 Aug 2020 13:08:29 UTC
All use subject to https://about.jstor.org/terms
 32 BENJAMIN W. LABAREE [AJA 61
wind conditions herecanas
be rated
wellas probable
assailing
over days, while
the 5.4 Bospo
On nearly twelve daysremain
in asthe
ideal. Theaverage
following table indicates sail-
sailing mon
a northeast wind blows over
ing conditions the
for the six upper
months of the sailing Black
area to assist the flow of water down toward the season:

Bosporus; whereas on ten days southwesterly winds Sailing inds AlAug. Sept.
actually tend to hold back the water from thisConditions Winds Apr. May June July Aug. Sept.
course. On the remaining days of the month the IDEAL Bosp.: southerly
winds are calm, thus not affecting this southwester- BI. Sea: southerly 7.4 7.3 5.4 3.0 2.6 3.5
PROBABLE Bosp.: southerly
ly drift.18
BI. Sea: northerlv 5.6 4-7 3.6 2.0 2.4 3.5
Winds at Istanbul are far more important, how-
IMPOss. Bosp.: northerly I7 19 21 269 26 23
ever. Not only do they directly affect the velocity
of current in the Bosporus, but they also constitute These figures represent as clearly as possible
the force upon which sailing vessels depend for "normal" conditions in the Bosporus over a six-
movement.? Southwesterly winds here not only month period. They do not take into account, how-
tend to check the current through the straits, but ever, unusual situations created by storms. For ex-
at the same time they become the favorable "fol- ample, a strong northeast storm over the Black Sea
lowing winds" which enable square-rigged vessels area will drive more than the normal amount of
to sail through the passage into the Black Sea be- water down through the Bosporus. As a result, for
yond.20 Such a wind in fact does blow on more the next few days thereafter, almost no adverse
than eight days during the average sailing month. current is to be found in the strait, because it takes
Sailing through the Bosporus depends, then, on a some time for the level of the Black Sea to rise
combination of wind conditions both in that strait again above that of the Sea of Marmara. If a north-
and over the Black Sea as a whole. With this knowl- east gale were to be followed immediately by fresh
edge we can determine both definite and probable southerly winds, a current might even flow north
sailing days for each month of the sailing season by through the Bosporus into the Black Sea."21 These
the following method: from the number of days are exceptional occurrences, however, and are men-
of southerly winds in the Bosporus (the essential tioned here mainly to show how important in this
condition for sailing through) subtract those days problem is the behavior of the wind.
during which it is probable that northerlies over These calculations based on observations taking
the Black Sea are increasing the current in the place in the late nineteenth century22 are not neces-
strait. Thus, if, as in the month of June, northerlies sarily applicable to the ancient period, with which
blow over the northern part of the Black Sea on we are primarily concerned here. And yet in
twelve of thirty days (or forty percent.) then on the absence of evidence to the contrary, we can
forty percent. of the "southwesterly days" at the only assume that conditions were basically the same
Bosporus currents can be expected to be of above- then as now. Carpenter cites one possible difference,
average velocity. Of the nine days in June with viz. that climatologists believe the first millennium
southwesterlies in the Bosporus, therefore, 3.6 days B.c. to have been unusually wet, and that therefore
18 The number of days each month during which winds inin Black Sea Pilot Ap
the northwest part of the Black Sea blow from the two prin- winds marked "north
cipal directions with which we are concerned has been de-the Bosporus, while
termined by averaging the results of wind observations at Var-There are no days of
na, Odessa, and Sevastopol (underlined on the accompanying Istanbul Apr. May June July Aug. Sept.
map). See Black Sea Pilot Appendix III, 426 (Varna) 428
(Odessa) and 430 (Sevastopol). In the following table, winds northerlies 17 19 21 26 26 23
marked "northerly" blow toward the Bosporus, those marked southerlies i 13 12 91 5 5 7
"southerly" away from this area:
21 ibid. 49.
Northwest Black Sea Apr. May June July Aug. Sept. 22 The meteorological tables to be found in Appendix III
northerlies 13 12 12 13 15 15 of the Black Sea Pilot were taken from the British Admiralty
Sailing Directions and are based on various periods of observa-
southerlies 12 13 13 12 11 10
tion as noted above each table. As in the case of the observa-
calm 5 6 5 6 5 5
tions on currents in the Bosporus (see footnote 14), it makes
19 ibid. 45, 49. little difference to the present study that these observations
20 The number of were not recently made. each
days month du
Istanbul blow from the two principal di

This content downloaded from

62.29.13.150 on Thu, 27 Aug 2020 13:08:29 UTC
All use subject to https://about.jstor.org/terms
1957] HOW THE GREEKS SAILED INTO THE BLACK SEA 33
the Black Sea received even more water than in Leucon ... has granted exemption from duty to
the modern period. This possibility, on the other
those who export to Athens ... When he made a
new harbor at Theodosia [in the Crimea] ... he
hand, is countered by the fact that centuries of cul-
tivation have stripped the natural vegetation fromgave us [Athenians] the exemption there also."24
the watersheds of the great rivers flowing into the To be sure, triremes and other galley-type ves-
Black Sea. One might conclude from this that sels a entered the Black Sea, but not as cargo car-
higher percentage of precipitation, unchecked by riers. Rather, as is suggested by Aristotle and De-
mosthenes, the role of the warship was to protect
forest and undergrowth, now drains into the rivers.
Whatever the precise amount of water flowing into the grain fleet from attack by the inhabitants of
the Black Sea in ancient times, the behavior of the
Byzantium and Chalcedonia.25
winds remains the basic factor affecting navigation One can easily see how these merchantmen
of the Bosporus. There is no evidence that winds passed through the tricky straits on their voyage
into the Black Sea. After several stops along the
in ancient times differed from the pattern observed
in the nineteenth and twentieth centuries. route, they arrived off the Bosporus prepared to
The strongest evidence from antiquity to wait
be for the appearance of a favorable wind to
take them through. In the spring, when Greek mer-
found influencing this question of penetration into
the Black Sea is the fact that the Greeks wrote chants would most likely send their vessels in for
the Black Sea grain crop, this delay might be ex-
about vessels sailing to the Pontus as if there were
no particular problem involved. Demosthenes' Ora- pected to last four or five days at most. In summer
a
tion Against Lacritus deals specifically with a law- week or more might pass before a southwest
suit arising from the following agreement: wind came up. With a good breeze from behind
and the current slowed to two knots, or perhaps
Androcles of Sphettus and Nausicrates of Carystus
have lent to Artemo and Apollodorus, both of Pha-less, the typical merchantman could negotiate the
seventeen-mile passage in eight or nine hours, un-
selis, three thousand drachms in silver from Athens
to Mende or Scione and thence to Bosporus [in the der sail alone. One may conclude, therefore, that it
Crimea], or, if they please, on the left coast as far
was regularly possible to sail through the Bosporus.
as the Borysthenes [the Dnieper River] and back
In no other way could the Greeks carry on an ac-
to Athens, at interest of two hundred and twenty-
tive trade with the Pontus.2
five for the thousand ... .23
In another oration Demosthenes notes that "... HARVARD UNIVERSITY

23 Demosthenes, Oration Against Lacritus. Greek settlements inside the Black Sea. Evidence from the
24 Demosthenes, Against the Law of Leptines. Black Sea Pilot seemed to support the theory that sailing through
25 Aristotle, Economicus, 2.1x; Demosthenes, the
50.6, 17 and
Bosporus was a difficult undertaking. His failure to realize
5.25. For alleged imperial Athenian development that
fromprevailing
a sea-winds do not necessarily blow every day of the
season, however,
route to the Black Sea as opposed to an alleged land-route to led him to his erroneous conclusion that at
the Hebros see ATL I, 465 and references. no time could a vessel sail through the Bosporus. His preoc-
26 Briefly, it would appear that Rhys Carpentercupation with the problem of conditions at the Bosporus ob-
went astray
in the following manner. Having determined that scured the more
it was the fundamental point that the pentekonters
pentekonter and not the trireme which was developedcouldby
notAmei-
successfully be used for trading purposes under any
circumstances.
nokles around 680 B.c., Carpenter decided to investigate whether
this date had any bearing on the problem of dating the earliest

This content downloaded from

62.29.13.150 on Thu, 27 Aug 2020 13:08:29 UTC
All use subject to https://about.jstor.org/terms