Skip to main content

Full text of "Pharsalia, Pharsalus, Palaepharsalus"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 


The site of the battle-field where Caesar finally defeated Pompey, 
on the ninth of August, 48 B. C, was held by Leake, Northern 
Greece, IV, p. 476 ff., to be between Pharsalus (modern Fersala) 
and the river Enipeus. This river he identified with the modern 
Tsanarli, although he gives it the modern name Fersalitis. He 
located the camp of Pompey on the heights east of Fersala. that 
of Caesar at the foot of the rocky height which advances into the 
plain three miles westward of Fersala. His locations and general 
plan of the battle were followed by Drumann, Geschichte Roms, 
III, p. 503 ff., and by Merivale, Hist, of the Romans, II, p. 227 ff. 
Merivale pointed out some difficulties in the way of Leake's views, 
which were commented on by the latter in a paper cited and 
summarized by the writer of the article " Pharsalus " in Smith's 
Diet, of Greek and Roman Geography. The essay itself I have 
been unable to consult. Leake's views, however, remained 

An entirely different location of the battle-field was made by 
Goler, Caesars Gallischer Krieg und Theile seines Burgerkriegs 2 , 
II, p. 150 ff., namely, on the further side from Pharsalus of the 
river called Enipeus by Leake. This river Goler calls Apidanus, 
and a very small tributary stream between Pharsalus and his 
Apidanus, entirely east of the road leading north from Pharsalus 
to Larissa, he calls Enipeus. A similar tributary appears without 
name on Leake's map, but it is made to rise a little to the west of 
Pharsalus, and to lie entirely west of the road to Larissa. Goler's 
plan of the battle-field is based on an Austrian military map, 
which is inaccessible to me, but must vary materially from the 
later maps of Kiepert. These are certainly more trustworthy. 
Caesar's camp is placed by Goler on the right bank of his 
Apidanus, just where the road from Larissa crosses the river; 
Pompey's camp is placed directly across the river valley, on the 
heights called Cynoscephalae. The main features of Goler's plan, 
but not his nomenclature, nor his minor departures from standard 
cartography, are adopted and ably defended by Long, Decline of 
the Roman Republic, V, p. 213 ff. 


Mommsen, Hist, of Rome, Am. ed., 1 IV, p. 495 if., follows Goler 
so far as to locate the camp of Pompey on the further side from 
Pharsalus of the Enipeus. This river Pompey's whole army 
crosses to give battle to Caesar. In all else Mommsen follows 
Leake. This combination is so peculiar that I quote in full all 
those parts of Mommsen 's description of the battle which involve 
it, or which are referred to in the following arguments. In the 
text we read : " Caesar lay to the south of Larissa in the plain — 
which extends between the hill-country of Cynoscephalae and the 
chain of Othrys and is intersected by a tributary of the Peneius, 
the Enipeus — on the left bank of the latter stream near the town 
of Pharsalus ; Pompeius pitched his camp opposite to him on the 
right bank of the Enipeus along the slope of the heights of 
Cynoscephalae." " When Pompeius hesitated as to his crossing of 
the rivulet which separated the two armies, and which Caesar with 
his much weaker army did not venture to pass, this excited great 
indignation." " Pompeius yielded ; and Caesar, who, under the 
impression that matters would not come to a battle, had just 
projected a mode of turning the enemy's army, and for that purpose 
was on the point of setting out towards Scotussa, likewise arrayed 
his legions for battle, when he saw the Pompeians preparing to 
offer it to him on his bank." " Pompeius rested his right wing on 
the Enipeus ; Caesar opposite to him rested his left on the broken 
ground stretching in front of the Enipeus ; the two other wings 
were stationed out in the plain, covered in each case by the cavalry 
and the light troops." Pompey's cavalry, attacked by Caesar's 
quarta acies, "galloped at full speed from the field of battle." 
"When Pompeius, who from the outset did not trust his infantry, 
saw the horsemen gallop off, he rode back at once from the field 
of battle to the camp, without even awaiting the issue of the 
general attack ordered by Caesar. His legions began to waver 
and soon to retire over the brook into the camp, which was not 
accomplished without severe loss." When Pompey " saw the 
legions retire over the stream he . . . rode off by the nearest route 
to the sea." In a footnote to the first sentence quoted from the 
main text Mommsen says : " The exact determination of the field 
of battle is difficult. Appian (II 75) expressly places it between 
(New) Pharsalus (now Fersala) and the Enipeus. Of the two 

1 This translation was compared with the fourth German edition. Citations 
from the sixth show that the author has made no changes in the portions which 
come under discussion here. 


streams, which alone are of any importance in the question, and 
are undoubtedly the Apidanus and Enipeus of the ancients — the 
Sofadhitiko and the Fersaliti — the former has its sources in the 
mountains of Thaumaci (Dhomoko) and the Dolopian heights, 
the latter in Mount Othrys, and the Fersaliti alone flows past 
Pharsalus ; now as the Enipeus, according to Strabo (IX, p. 432). 
springs from Mount Othrys and flows past Pharsalus, the Fersaliti 
has been most justly pronounced by Leake (Northern Greece, IV 
320) to be the Enipeus, and the hypothesis followed by Goler that 
the Fersaliti is the Apidanus is untenable. With this all the other 
statements of the ancients as to the two rivers agree. Only we 
must doubtless assume with Leake, that the river Vlokho, formed 
by the union of the Fersaliti and the Sofadhitiko and going to the 
Peneius, was called by the ancients Apidanus as well as the Sofa- 
dhitiko ; which, however, is the more natural, as while the Sofadhitiko 
probably has, the Fersaliti has not, constantly water (Leake, IV 321). 
Old Pharsalus, from which the battle takes its name, must therefore 
have been situated between Fersala and the Fersaliti. Accordingly 
the battle was fought on the left bank of the Fersaliti, and in such 
a way that the Pompeians, standing with their faces towards 
Pharsalus, leaned their right wing on the river (Caesar, B. C. Ill 83 ; 
Frontinus, Strat. II 3, 22). The camp of the Pompeians, however, 
cannot have stood here, but only on the slope of the heights of 
Cynoscephalae, on the right bank of the Enipeus, partly because 
they barred the route of Caesar to Scotussa, partly because their 
line of retreat evidently went over the mountains above the camp 
towards Larissa; if they had, according to Leake's hypothesis 
(IV 482), encamped to the east of Pharsalus on the left bank of the 
Enipeus, they could never have got to the northward through this 
stream, which at this very p.oint has a deeply cut bed (Leake, IV 
469), and Pompeius must have fled to Lamia instead of Larissa. 
Probably, therefore, the Pompeians pitched their camp on the 
right bank of the Fersaliti, and passed the river both in order to 
fight and in order, after the battle, to regain their camp, whence 
they then moved up the slopes of Crannon and Scotussa, which 
culminate above the latter place in the heights of Cynoscephalae. 
This was not impossible. The Enipeus is a small slow-flowing 
rivulet, which Leake found two feet deep in November, and which 
in the hot season often lies quite dry (Leake, I 448, and IV 472 ; 
cf. Luc. VI 373), and the battle was fought in the height of summer. 
Further, the armies before the battle lay three miles and a half 


from each other (Appian, B. C. II 65), so that the Pompeians could 
make all preparations and also properly secure the communication 
with their camp by bridges. Had the battle terminated in a 
complete rout, no doubt the retreat to and over the river could 
not have been executed, and doubtless for this reason Pompeius 
only reluctantly agreed to fight here. The left wing of the 
Pompeians, which was the most remote from the base of retreat, 
felt this ; but the retreat at least of their centre and their right 
wing was not accomplished in such haste as to be impracticable 
under the given conditions. Caesar and his copyists are silent as 
to the crossing of the river, because this would place in too clear 
a light the eagerness 'for battle of the Pompeians, apparent 
otherwise from the whole narrative, and they are also silent as to 
the conditions of retreat favorable for these." 

To this apparent combination of the views of Leake and Goler 
by Mommsen, Goler objected, on the ground principally that the 
same reasons which forbade locating Pompey's camp on the left 
bank of the Apidanus (Enipeus) forbade also making Pompey's 
army cross the river to give battle, and recross it in flight. 

In all these authorities the names of the modern rivers corre- 
sponding to those anciently called Apidanus and Enipeus differ 
from those of the newest maps of Thessaly, and any one who 
compares the successive editions of Kiepert's classical maps of 
Greece, large and small, will find surprising vacillation in the nexus 
and nomenclature of all the streams draining the great watershed 
of Thessaliotis. But since the cession of Thessaly to the kingdom 
of Greece in 1881 made scientific surveys necessary, and rendered 
travel and investigation in this district safe and inviting, many 
topographical questions hitherto doubtful have been finally settled, 
and much light has been shed on the conflicting statements of 
ancient writers regarding the main geographical features of 
Thessaly. It seems worth while, therefore, to review the question 
of the river system of the Pharsalian plains, and the site of 
Caesar's greatest battle with Pompey, in the light of the latest 
maps. 1 Such a review seems called for also from such facts as 
these, that Badeker's most welcome handbook for travellers in 
Greece presents routes in Thessaly on the basis of Kiepert's latest 
map, but, apropos of Fersala, quotes Mommsen's account of the 

1 H. Kiepert, Atlas Antiquus, No. 5 and 6, edition of i832; Karte des 
Konigreichs Hellas, accompanying Badeker's Griechenland. 


battle of Pharsalus ; and that the last editor of Appian refers to 
this as the standard account. This account, however, was written 
before the geography of the Pharsalian plain had become definitely 
fixed, and would doubtless be materially changed should the 
eminent historian revise again the volume in which it is contained. 
It will not therefore be thought presumptuous in me to present 
some criticisms of this account. 1 The "historical artisan" may 
sometimes properly criticize details in the work of the " historical 

Of the five main rivers of Thessaly mentioned in Hdt. VII 129, 
four, the Pamisus, the Onochonus, the Apidanus, and the Enipeus, 
are now described by Kiepert as flowing into the fifth, the Peneius, 
from the south, and as draining Thessaliotis in the order mentioned 
from west to east. Of these, two, the Apidanus and Enipeus, 
passed through the Pharsalian plains, but just how has been until 
recently quite uncertain, and very differently represented on 
different maps. In VII 196, Herodotus says that the Onochonus 
was the only river of Thessaly which could not supply the army of 
Xerxes with water, as if this were the smallest of the tributaries of 
the Peneius. 2 But he adds that of the rivers of Achaia, even the 
largest, the Apidanus, fared almost as badly as the Onochonus. 
The Apidanus therefore, according to Herodotus, was a large river 
in both Thessaly (Thessaliotis) and Achaia (Phthiotis). No river 
corresponds to this description except that called the Enipeus by 
Kiepert, rising in the Othrys range of Achaia, taking a north- 
easterly and northerly course through Achaia, then a north-westerly 
along the north-eastern side of Thessaliotis. Without attempting 
to notice the explanations of this last statement of Herodotus 
which have been made, it is enough for my present purpose to 
say that it is now clear that he confounded the Apidanus with the 

With regard to these two rivers Thukydides, IV 78, is perfectly 
accurate. When Brasidas, after the battle of Delium, attempted 

1 Most of this paper was already in manuscript when I received Seldner's 
Schlachtfeld von Pharsalus, a Mannheim school-program of 1882-3. Many 
of my objections to Mommsen's account of the battle I find anticipated here, 
but the program was written before Kiepert's last maps of Thessaly were 
published, and the method of the enquiry will be seen to be quite different 
from mine. 

2 The schol. on Apoll. Rh. Ill 10S5, in quoting Hdt. VII 119, omits the 
Onochonus entirely. 


to conduct an armed force through Thessaly into Macedonia, the 
popular sentiment of Thessaly was against him, and an opposing 
force stopped him at the Enipeus, in Achaia (Phthiotis), just above 
Melitia. At this point, therefore, the crossing of the Enipeus was 
strategically important. Having so far satisfied his opponents of 
his intentions that they dispersed, Brasidas, following the advice 
of his oligarchical friends and guides, stole his way through the 
country by forced marches into Perrhaebia, avoiding, of course, 
the large cities, and the main route by way of Larissa. Even on 

the day of the parley § ex rijr MeXin'aj a(pa>pp.r]<rev, e's &apo-a\6v re 
eVe'Xeo-e Km co-TpaTcmtbeio-tiTO em t£ 'Awidavin 7roTa/j.a. This river then, 

as Classen's note ad loc. correctly states, flowed at some distance 
to the north of Pharsalus, and the inference is a very probable one 
that it had not lain in the path of Brasidas since his crossing of 
the Enipeus in Achaia (Phthiotis).' The great river of Achaia 
Phthiotis as well as of Thessaliotis was therefore the Enipeus, 
and the Apidanus must have risen in the extreme southern slopes 
bounding the Pharsalian plain, near Pharsalus itself. So Kiepert 
now represents it. The Apidanus and Enipeus did not join, 
therefore, east or south-east of Pharsalus, as has been represented 
even by Kiepert in his earlier maps. But though the Enipeus 
was by far the'longer of the two rivers, its course before entering 
the great valley of Thessaliotis was through a mountainous 
country, and the Apidanus may well have been of equal or even 
greater volume at times, being fed by copious springs at the head 
of the valley about Pharsalus, and flowing through an almost 
marshy plain. After the junction of the two rivers, well towards 
the northern part of Thessaliotis and the Peneius (cf. Apoll. Rh. 
I 37 ff-)> tne united streams may have been variously called 
Apidanus or Enipeus, and the first name even erroneously extended 
to the Enipeus above the junction. It is otherwise hard to account 
for the confounding of the rivers in Herodotus, and for Strabo's 
statement, IX, p. 432, that the Enipeus flows from Othrys past 
Pharsalus, empties into the Apidanus, and this into the Peneius 

(o 6" EcMreit otto ttjs "Odpvos napa iapcraKov pvels els top 'Am&apov irapa- 

/3uXX«, o 8' els to iJriveiip). But a gloss (so Meineke) at Strabo VIII, 
p. 356, speaks of the Thessalian Enipeus as flowing from Othrys 

1 The route least likely to meet with further opposition would seem to have 
been through Thessaliotis on the west side of the Enipeus. and perhaps the 
uncertain Phakion, where Brasidas encamped after leaving the Apidanus, is to 
be located somewhere on this line, rather than in Pelasgiotis. 


and receiving the Apidanus after it has come down from Pharsalus 

(tov 5 iv tt) QerraXla EXnrea ypa<pov<rw, os affb rrjs v OBpvos picov Several 
top 'AmSavov Karevexdevra ex Qapcrakov). Whether this description IS 

Strabo's or not, it is certainly good evidence that the two streams 
at their junction were so nearly of a size as to make it doubtful 
which name the united streams should bear. 1 And however 
Strabo estimated the relative size of the rivers, he speaks of the 
Enipeus (VIII, p. 432) as near Melitaea, just as Thukydides does, 
and must therefore regard it as the longer of the two. 

In X 239 the Enipeus is called the most beautiful river in the 
world. It certainly was the most prominent tributary of the 
Peneius so far as length of course is considered, if, as the best line 
of evidence in ancient writers shows, and as Kiepert now decides, 
it was the eastern stream of Thessaliotis and Pharsalia, with the 
Apidanus next west. Moreover, at a point in Pharsalia between 
Pharsalus and Larissa, the Enipeus must have been the larger 
stream, while the Apidanus must have just begun its course, 
increasing so as to be of equal or even greater size where the two 
united. In any of the current locations of the battle-field, there- 
fore, the Enipeus will be the main river of the scene, whatever its 
special strategical importance may have been. So Lucan thinks 
of it when he prophesies (Phars. VII 116): Sanguine Romano 
quam turbidus ibit Enipeus, and, possibly with no more definite 
geographical purpose (v. 224), At iuxta fluvios, et stagna undantis 
Enipei, ) Cappadocum montana cohors, etc. The modern stream 
corresponding to the ancient Enipeus is now seen to be not the 
Fersalitis, but the Tsanarli. The modern Fersalitis is the ancient 
Apidanus. In all discussions of the site of the great battle between 
Caesar and Pompey the writer's standpoint regarding these streams 
of Pharsalia must first be made definite and clear. How perplexing 
the confusion of the two rivers is may be seen from reading the 
chapter on the battle-field in Long or Willmann (Adnotationes 
quaedam ad C. Julii Caesaris relationem pugnae Pharsalicae. 
Halberstadii, 1875). 

1 When Euripides calls the Apidanus a river of Phthia (Hec. 451, Iph. A. 
713), we may suppose that he adopts the geography of the heroic age, when 
Phthia included much if not all of Thessaliotis. Still Euripides gives much 
the same praise to the Apidanus that Homer does to the Enipeus, and it is quite 
possible that he interchanged the names, like Herodotus. The nomenclature 
of Herodotus probably lies at the basis of Apoll. Rh. II 514 f. Different 
descriptions of these same rivers in Ovid (Met. I 579 f.) and Lucan (Phars. VI 
372 f.) are undoubtedly due to an interchange of names. 


Pharsalus, the city which dominated the fertile territory so amply 
watered by these rivers, appears in Greek history under this name 
as early as 454 B. C. (Thuk. I in). It is even then a strong 
citadel and a representative Thessalian city, badly governed by a 
large landed class whose fighting force was cavalry. It continued 
such down to the time when Thessaly came under Roman control 
in 196 B. C. (Polyb. XVIII, 46 ff.). Of Leake's view that it was the 
Phthia of Peleus and Achilles, the capital city of Phthiotis in the 
Homeric age, there is no need to speak here, except as it empha- 
sizes the importance of the place. That' the city itself laid claim 
at least to Homeric antiquity may be reasonably inferred from the 
fact that it erected statues of Achilles and Patroclus at Delphi 
(Paus. X 13, 5). It is often mentioned by Thukydides, Xenophon, 
Polybius, Diodorus, Livy, and Plutarch. It was the only city of 
Thessaly which the Romans allowed to retain its freedom 
(Mommsen, Rom. Gesch. V, p. 273). 

But neither of this famous city nor of the famous river which 
flowed through its territory does Caesar make any mention 
whatever in his account of the batde. This is certainly remark- 
able if the battle was fought on a bank of the river or near 
the city. Their names could not be more significantly ignored if 
the struggle occurred miles away from them. Moreover, in 
speaking of the battle afterwards, he gives it no specific name. 
Twice he calls it proelium in T he ss alia factum. (B. C. Ill 101, 5; 
in, 3), the most indefinite expression possible for him to use. 
Cicero uses the same expression for the locality of the battle 
thrice (Phil. II 59; 75; de Divin. II 114). Thessalia may 
possibly denote here the district Thessaliotis, and so be somewhat 
narrowed down in content, but this is not Caesar's customary use 
of the word. The scene of battle is, however, very much narrowed 
down by the use of the adjective Pharsalian. Cicero speaks twice 
of the pugna Pharsalia, twice of the fuga Pharsalia, twice of 
the proelium Pharsalicum, twice of the acies Pharsalica. This 
last expression he uses in Caesar's presence (pro Ligario 9), and 
in addressing Antonius (Phil. II 71), who commanded the left wing 
for Caesar in the battle. Designations of the battle which consign 
it to the region about Pharsalus occur also in the historians so 
commonly as not to need citing. 

But a still more precise designation of the battle is found in 
Bell. Alex. 48, 1 (probably by Aulus Hirtius, the friend and officer 
of Caesar, though not himself present at the battle) : lis autem 


temporibus, quibus Caesar ad Dyrrhachium Pompeium obsidebat 
et Palaepharsali rem feliciter gerebat Alexandriaeque cum periculo 
magno, turn etiam maiore periculi fama dimicabat, etc. Here are 
chosen and exact expressions of locality, " near Dyrrhachium, at 
Palaepharsalus, at Alexandria." This narrower designation of the 
site of the battle is found also in Strabo (XVII, p. 796 : <=V tovtg> 

Hofi7rfjios Mdyvos t]kz (pevycav €K Ua\at<papo~d\ov irpos to Htj\ovo~ioi> kol to 

Kdo-iov opos), Frontinus (Strateg. II 3, 22 : Cn. Pompeius adversus 
C. Caesarem, Palaepharsali triplicem instruxit aciem), and Orosius 
(VI 15 : Hie exitus pugnae ad Palaeopharsalum fuit), the last two 
ultimately, if not directly, dependent on Livy. The origination 
and late survival of this exact designation among the far more 
numerous and easy general designations, tend to establish its 
correctness. If, then, the site of Palaepharsalus can be satisfactorily 
fixed, the site of the battle-field follows ; and if the site of the 
batde-field can be fixed, the site of Palaepharsalus follows. 
Neither can be done with absolute certainty ; but the evidence in 
both lines of enquiry points with strong probability to the same 
general locality. 

Col. Leake's opinion that Palaepharsalus was the ancient citadel 
just back of the modern Fersala, or that it was within very short 
distance of Pharsalus toward the Enipeus, must fall before the 
precision with which Strabo distinguishes the two places and uses 
each as limit of measure. Indeed, that the city and citadel of 
Pharsalus were distinct in ancient times as well as modern, as was 
the case elsewhere, may fairly be inferred from Xen. Hell. VI 1, 2, 
18, where the acropolis is said to have been reserved by Polydamas, 
but the city joins Jason of Pherae. In discussing the question 
whether the Homeric Hellas and Phthia were one (IX, p. 431), Strabo 
mentions as one opinion current that Hellas was not a city but a dis- 
trict, extending [«y] Tas Brjfias Tas &6ia>n8as airb IlaXaKpapo-dXov (cv Be Trj 
X&pq Tavrji Kai to Qeridciov eori TrXrjaiov to>v $apo~d\a>v dfKpoiv tt)s t€ nakaias 
Ka\ Ttjs veas, K€K tov GeriSci'ou TiKp.aipdp.evoi rijs tVo rffl 'A^tXX« pepos dvai 
Kal rr/vde ttjv x^>P av )i KT ^- The phrase likrio-lov . . veas has no 

particular force, and can with difficulty be accounted for if the 
two Pharsali were close to each other, or if either was very much 
nearer than the other to the Thetidium, or on the same line with it 
as the other. It is most naturally accounted for if Palaepharsalus 
and Pharsalus were approximately equidistant from the Thetidium. 
In that case, as Pharsalus lay at the extreme southern edge of the 
Pharsalian plains, Palaepharsalus would naturally be looked for 


toward the north or north-east. Strabo's language favors rather 
than forbids placing Palaepharsalus on the right of the Enipeus, 
inasmuch as the river valley would be a more natural boundary 
than a city within it. No indication is here given of the site of the 
Thetidium, further than that it was in the region between Palae- 
pharsalus and Phthiotio Thebes. 

The approximate site of the Thetidium, however, we get from 
Polybius, XVIII 20 (cf. Eurip. Androm. 16 if.). From the vicinity of 
Pherae Philip and Flamininus march by circuitous north and south 
routes to the westward, each trying to reach Scotussa first. Titus 
encamped the first night at Eretria in Phthia, Philip at the Onchestus. 
On the second night Philip encamped «Vl to MeXd^iov Trpoauyopeio^vov 

rrjs 2KOTV<raaias, and TitUS irepi to ©frt'Seiov rrjs &ap<ra\ias. The 

Thetidium was therefore in the territory of Pharsalia, which could 
not have extended far beyond the valley of the Enipeus, on the 
right of the Enipeus, on a line running south of Scotussa from 
Pherae westward, and on a military route between Eretria and 
Scotussa. These details enable us to locate it NE. of Pharsalus, 
nearly if not exactly where Col. Leake identified it with ruins then 
visible, and probably a little east of where Kiepert now locates 
Palaepharsalus. This latter, to accord best with our inference from 
Strabo that it was practically equidistant with Pharsalus from the 
Thetidium, and was held by some to be the western limit of Hellas, 
we must locate further to the NW., and even across the main 
route between Larissa and Pharsalus. 

To the same location we are led by general military considerations 
based upon the previous progress of the campaign between Caesar 
and Pompey, and by a proper interpretation of the language of 
Caesar. This ground has been thoroughly worked by Goler and 
Long. I can add but little to their arguments. After Caesar's 
great defeat at Dyrrhachium he retired into Thessaly, for the 
purpose of restoring the confidence of his men and securing ample 
supplies (B. C. Ill 74, 3). On the way he made a lucky junction 
with his officer, Domitius Calvinus, near Aeginium, the last town 
of Epirus in the upper valley of the Peneius (c. 79). He had 
previously (c. 34) sent fifteen cohorts and over two hundred 
cavalry into Thessaly and Aetolia, and subsequently ordered them 
further south into Achaia. They had been kept back, however, 
by an officer of Pompey at the Isthmus of Corinth, and were now 
engaged in winning Boeotia over to Caesar (c. 55). Pompey was 
known to be on his way through Macedonia to join Scipio at 


Larissa in Thessaly (cc. 79, 80, 82). Accordingly, in view of the 
great numerical inferiority of Caesar's forces, the first requirement 
of good generalship on his part would be to take up such a 
position in Thessaly, south of Larissa, as would put him in 
communication with his forces in Boeotia (cf. Plut. Caes. 43), 
prevent Pompey from reaching and crushing them, and at the 
same time command a large and fertile share of the Thessalian 
plains. Just such a place would be the road from Larissa to 
Pharsalus, where it leaves the low range of hills dividing Pelasgiotis 
from Thessaliotis, and enters the plain of the Enipeus and the 
territory of Pharsalus. 

What indications Caesar gives of his route through Thessaly 
point to this locality. He sacks Gomphi for closing its gates upon 
him, and spares Metropolis because it receives him. Thereupon, 
nulla Thessaliae fuit civitas praeter Larisaeos, qui magnis exercitibus 
Scipionis tenebantur, quin Caesari parerent atque imperata facerent 
(c. 81). This certainly includes Pharsalus, and made it unnecessary 
for Caesar to visit it, as well as quite improbable that the battle 
should subsequently be fought under its walls without any notice 
being taken by Caesar of the city and its attitude. Caesar's 
description of his course after leaving Metropolis is vague, mainly 
because it took him to no city, and to no place easily designated 
by its special nearness to any city. Hie ' idoneum locum in agris 
nactus . . . ibi adventum exspectare Pompei eoque omnem belli 
rationem conferre constituit (c. 81, 3). This place must have been 
" suitable " not only for controlling a large area of the ripening 
harvest, but also for awaiting Pompey's advance southwards. It 
could not therefore, as all military critics say at once, have been at 
a point which would have left the great road south from Pharsalus 
open to Pompey, and Caesar had time to make deliberate choice. 
It was not near enough to either the Enipeus or Pharsalus to bring 
them into special mention. 

Pompey did not effect a junction with Scipio until some days 
after Caesar had established himself in this position (c. 82, 1). 
Caesar says nothing of the advance of these united forces from 
Larissa, nor does he locate clearly the camp they occupied just 
before the battle. He gives, however, some significant hints in 
his brief description of the battle, all of which point to the southern 
slope of the range of hills dividing Pelasgiotis and Thessaliotis, 

1 1 regard it as now beyond controversy that this pronoun refers to Caesar 
and not to Scipio. See Willmann, /. c. p. 3 f. 


near the great north and south route running from Larissa to 
Pharsalus. The camp of Pompey must have been determined 
by that of Caesar, and over against it, since Caesar was now on 
the defensive, and Pompey at last on the aggressive, driven on by 
the impatience of the senatorial party and their overestimate of 
the victory at Dyrrhachium. But, after encamping over against 
Caesar, Pompey's old caution returned, and he kept deferring his 
attack until Caesar determined even to challenge him (c. 84, 1). 
Itaque ex castris exercitum eduxit aciemque instruxit, primo suis 
locis pauloque a castris Pompei longius, continentibus vero diebus, 
ut progrederetur a castris suis collibusque Pompeianis aciem 
subiceret (c. 84, 2). From this it may be inferred that considerable 
distance intervened between the camps, and that Caesar's was on 
low ground compared with Pompey's, especially as Caesar so 
particularizes this contrast by speaking of his own as in agris 
(c. 81, 3), and by saying (c. 85, 1): Pompeius, qui castra in colle 
habebat, as though this was an advantage on Pompey's side. At 
last, just as Caesar is about to break camp and enter on a flying 
campaign, in despair of bringing matters to a crisis, Pompey's 
forces come so far out into the plain from their high camp that a 
battle can be fought non iniquo loco (c. 85, 3). Caesar still 
advances further to the attack (c. 88, 1), and Pompey's line await 
his charge (c. 92, 2). 

When Pompey's great body of cavalry had been routed by 
Caesar's famous quarta acies, they all turned, and not only 
abandoned the field, but without stopping, fled at the top of their 
speed to the highest hills (omnesque conversi non solum loco 
excederent, sed protinus incitati fuga montes altissimos peterent ; 
c 93, 6). There were, then, in the rear of Pompey's left wing 
high hills. That these were not at right angles to his line of battle 
is clear from the fact that the same charge which routed his cavalry 
brought Caesar's quarta acies upon the left of his infantry line 
(eodem impetu cohortes sinistrum cornu pugnantibus etiam turn 
ac resistentibus in acie Pompeianis circumierunt eosque a tergo 
sunt adorti ; c. 93, 8). Moreover, that these hills were part of the 
range toward the base of which Pompey's camp was pitched, is 
probable from Caesar's description of his storming the camp (c. 95). 
After a feeble resistance the garrison withdrew from the defences, 
protinusque omnes ducibus usi centurionibus tribunisque militum 
in altissimos montes, qui ad castra pertinebant, confugerunt. 
Pompey's force was so large that either wing of the battle line 


would project beyond the camp, and the horsemen, on the left 
wing, took the bee-line of panic-flight past even the camp. 

Pompey, finding the enemy in his camp, rode out of the decuman 
gate, and without stopping made at full speed for Larissa (decumana 
porta se ex castris eiecit protinusque equo citato Larisam contendit ; 
c- 96, 3). The most natural inference from this passage is certainly 
that the decuman or rear gate looked toward Larissa. With 
Caesar's cavalry scouring the country Pompey could not take a 
roundabout course. Pompey had advanced southward from 
Larissa until he confronted Caesar, and had then entrenched 
himself on the hills sloping down into the Pharsalian plains. 
Nothing but the most forced explanation can make the passage 
consist with Leake's position for Pompey's camp and line of battle, 
facing north on the plain just east of Pharsalus. This difficulty 
Merivale and Mommsen recognize fully. 

The difficulty is duplicated by Caesar's statement that Pompey's 
soldiers, after fleeing to a position in the hills back of the camp, 
abandoned it on seeing Caesar preparing to blockade it, since it 
had no water, and started along the mountain ridges toward 
Larissa (relicto monte universi iugis eius Larisam versus se recipere 
coeperunt ; c. 97, 2). It would have been a hopeless undertaking 
to reach Larissa from Leake's position for Pompey's camp, while 
Caesar held the main road. As it was, Caesar took a better route 
(probably the main road between Pharsalus and Larissa), and 
headed off the retreating crowd after a march of six miles 
(commodioreque itinere Pompeianis occurrere coepit, et progressus 
milia passuum sex aciem instruxit ; c. 97, 3). This was at a hill 
whose base was washed by a certain river (Hunc montem flumen 
subluebat ; c. 97, 4). The only hill which Leake could find 
answering to this description was near Scotussa, and washed by the 
Onochonus (now called the Onchestus). But Leake admits frankly 
that it was more than six miles from the banks of the Enipeus. 
" If we suppose Caesar to have computed his distance of six miles 
from the banks of the Enipeus north-eastward of Fersala, and to 
have encamped at some little distance short of the Onochonus, the 
march would not have been much greater than six miles, though 
it seems rather underrated at this distance." Adopting, then, 
Leake's identification of this hill (and nothing seems improbable 
in it), and Caesar's march of six miles was reckoned rather from 
the northern edge of the broad valley of the Enipeus, where our 
enquiry thus far tends to place the camp of Pompey, than from 


the river itself, to say nothing of the southern side of the valley 
where Leake and those who follow him locate the battle-field. As 
Pompey's camp, or rather an eminence in the rear of it, was the 
starting point of the march, there is no good reason for including 
the plain between this and the Enipeus in Caesar's estimate of the 
length of the march. The new maps show no mountain nearer 
the valley of the Enipeus than the one which Leake fixes upon, 
though they suggest the identification of some one of the hills 
more in the direction of Larissa, washed by tributaries of the 
upper Onchestus, with the hill so vaguely described by Caesar. 
From this hill, on the following day, after receiving the surrender 
of the beleaguered enemy, and after bringing up relief legions 
from his old camp, Caesar proceeds to Larissa (c. 98, 3). This 
implies the very close proximity of this city to the scene of the 

The argument from Caesar's Commentaries, drawn from 
incidental and indirect allusions to the geography of the field of 
battle, is cumulative in establishing the probability that the camps 
both of Pompey and Caesar were on the side of the Enipeus 
toward Larissa, and that the camp of Pompey was on the southern 
slope of the hills bounding the northern edge of the Pharsalian 
plain. Such positions are also demanded by the most general 
military considerations. In this neighborhood, too, Palaepharsalus, 
the stricter designation of the locality of the battle first found in 
Aulus Hirtius and surviving even to Orosius, is best located. 

This name occurs in Strabo, as we have seen, without designating 
the site of the battle between Caesar and Pompey. Livy also uses 
it in careful distinction from Pharsalus. During the desultory 
third Macedonian war the Roman consul for a long time held a 
position in Thessaly from which he hoped to advance against 
Perseus, strongly entrenched on the confines of Macedonia. He 
was of course also liable to attack from Perseus. The same 
general military considerations as in the case of Caesar's campaign 
in Thessaly would lead him to occupy such a position as would 
command the fertile Pharsalian plains, and control the great north 
and south route through Thessaly. Livy says (44, 1) : castra eo 
tempore A. Hostilius in Thessalia circa Palaepharsalum habebat. 

Long gives (/. c. p. 220) from private correspondence General 
Sir Wm. Napier's objections to the site of the batde as designated 
by Leake. " It seems impossible that a great general like Caesar 
should allow Pompeius to pass the Enipeus before him and cut 


him off from Pharsalus and Scotussa, and also from one of the 
roads to Thermopylae, which endangered Caesar's troops in 
Greece. It is also impossible that so great a general as Pompeius 
would pass the Enipeus in the face of Caesar's army, leaving his 
own place of arms, Larissa, open to his enemy ; moreover, Caesar 
does not mention Pompey's passage of the river. He does not 
indeed mention his own, but there was no need of that ; it was 
part of his march when no enemy was near him." These objections 
are sustained by the whole course of our enquiry thus far. 

Against Mommsen's peculiar modification of Leake's view, that 
Pompey's camp was on the north side of the Enipeus, but that he 
crossed the Enipeus to attack Caesar, that his cavalry recrossed 
it in their fatal flight to the hills, and that his whole army recrossed 
it to regain their camp after their defeat, I note the following 
additional points which have not been already brought out 
explicitly in the course of the enquiry. First, no mention of such 
crossing and recrossing is even implied in any ancient authority 
for the battle, although it must have formed a very important 
feature of the struggle. Second, the assumption that the battle 
did not terminate in a complete rout is also contrary to all the 
evidence we have, and if this be distrusted as too partisan, to the 
undisputed and indisputable fact that Pompey's camp was taken 
by storm. Caesar says of the troops which formed Pompey's 
main line of battle, after the attack of the tertia acies, universi 
terga verterunt (c. 94, 2), initium fugae factum (c. 94, 4), Pompeianis 
ex fuga intra vallum compulsis . . . perterritis (c. 95, 1), qui ex 
acie refugerant milites (c. 95, 4). Third, the motive assigned for 
the silence of " Caesar and his copyists" about Pompey's thus 
crossing the river, that it would place in too clear a light his 
eagerness for battle, is not only insufficient if it could be shown to 
exist, but is absolutely precluded. Caesar dwells upon the eager- 
ness of the Pompeians to fight him (cc. 82, 83, 86, 87) in 
consequence of his defeat at Dyrrhachium. He makes the military 
caution of Pompey himself the only restraining element. It was 
this which led Pompey to await attack from Caesar, instead of 
advancing to give it (cf. c. 85 ; 88, 1). Moreover, there were some 
in the senatorial party who dreaded the battle and feared the 
result, so that Pompey must have had some support in his cautious 
procedure, if reliance .can be placed on the testimony of the 
vacillating Cicero (de Divin. II 114): Hie [rem ex] vero et ea 
quidem [praedixit] quae omnes eo tempore ne acciderent timebamus 


. . . Videbaturque nobis exercitus Caesaris et audaciae plus habere, 
quippe qui patriae bellum intulisset, et roboris propter vetustatem. 
Casum autem proelii nemo nostrum erat quin timeret. 

No historian now holds Arnold of Rugby's contempt for Caesar's 
Commentaries on the Civil War as an authority, least of all M ommsen. 
Even granting that Caesar colors his accounts of political measures 
in his own favor, his descriptions of purely military operations 
will stand every test of historical fidelity. No writer has described 
the defeat at Dyrrhachium in darker colors than he. No one 
certainly was better able to describe the battle of Palaepharsalus. 
What unintentional indications of the site of the battle we get from 
his brief description of the purely military features of the struggle 
are of the highest value. They all bear, I venture to think, 
against the views of Mommsen. 

Regarding it as proven, then, that the battle of Palaepharsalus 
was fought on the north of the Enipeus where the camps of both 
armies had been pitched, and that Pompey's camp at least was on 
the hills sloping toward the river valley, I shall briefly notice 
another question, the more exact location of Caesar's camp. Goler 
places it at the crossing of the Enipeus by the road between 
Pharsalus and Larissa, which it secured, and opposite Palaepharsalus, 
a league distant upon the hills. He cites as a similar procedure 
of Caesar's, B. G. II 5. Sir William Napier, quoted by Long (7. c. 
p. 221), places Caesar with Scotussa in his rear, and his camp, 
of course, facing west. He places the camp of Pompey facing 
the east at the foot of some heights which border the Enipeus. 
We have seen that Caesar's language implies by way of contrast 
that his own camp was in the plain, aside from his vague 
expression "locum in agris." With this restriction, Napier's 
location can be defended, but no very positive preference of his 
view or that of Goler can be justified with the evidence now at 

Speaking of Pompey's line of battle, Caesar says (c. 88, 6) : 
Dextrum cornu eius rivus quidam impeditis ripis muniebat. Can 
this " rivus quidam " be the principal river of Thessaliotis, the 
divine Enipeus, bs iro\v KaWurros myrafiSiv M yalav <ijaw? Merivale 
recognizes the difficulty in the way of this identification, and calls 
such a use of rivus instead of Jlumen " against Caesar's and all 
correct usage." I may add that to a stream so insignificant that it 
has not yet been conclusively identified at all, and can in no case 


be one of the main rivers of Thessaly, 1 Caesar gives the name 
flumen (c. 97, 4). It is also dangerous to argue from the insigni- 
ficant volume of rivers in Greece to-day, that they were equally 
insignificant twenty centuries ago. This Leake and those who 
follow him have done. The rivus quidam must have been, as 
Goler and Long argue, one of the many mountain streams flowing 
down from the hills between Pelasgiotis and Thessaliotis into the 
Enipeus. Two such streams are represented on Kiepert's last 
maps of ancient and modern Greece (not identically in both), one 
on each side of the main road from Pharsalus to Larissa. 

Against this view are the following ancient authorities, the relative 
value of whose testimony must now be considered: Frontinus 
and Orosius, in what they say of the battle (see p. 178), are 
generally supposed to have drawn from Livy's lost one hundred 
and eleventh book. 2 As they both locate the battle at Palaephar- 
salus, it is probable that Livy did so. 3 Frontinus, however, has 
another statement which would make it appear that Livy called 
the stream which covered Pompey's right, the " rivus quidam 
impeditis ripis " of Caesar's Commentaries, the Enipeus : dextro 
latere [conlocavit] sexcentos equites propter flumen Enipea, qui 
et alveo suo et alluvie regionem impedierat. Orosius follows this 
version in so far as he has Pompey station a small body of horse- 
men on his right (in dextro quingenti). This is in conflict with 
Caesar (c. 88, 6) : quam obcausam cunctum equitatum, sagittarios 
funditoresque omnes sinistra cornu obiecerat. Orosius does not 
state the reason why Pompey put so few cavalry on his right, and 
so, of course, does not mention the stream which covered that 

In still another point was Livy (as represented by Orosius) at 
variance with Caesar, viz. in the number of troops engaged on 
both sides. Pompey's forces Caesar gives (c. 88, 5) as one 
hundred and ten cohorts, or forty-five thousand regular legionary 
soldiers, besides two thousand evocati. Pompey's cavalry Caesar 
estimates at seven thousand, against his own one thousand (c. 84, 
4). But Livy (Orosius) puts Pompey's line of battle at eighty- 
eight cohorts only, or forty thousand men, and his cavalry, on both 

1 On what evidence Drumann concludes that the stream at the base of the 
hill on which the remnants of Pompey's army made their final stand was the 
Enipeus (/. c. p. 515), I cannot imagine. 

2 Bludau, de fontibns Frontini, Diss. Regimont, Brunsbergae, 1883. 

3 The Epitome, however, has " apud Pharsaliam." 


wings together, at only eleven hundred, praeterea reges multi, 
senatores, equitesque Romani plurimi absque levium armatorum 
magna copia. Again, Caesar states his own force engaged to have 
consisted of eighty cohorts, or twenty-two thousand men (c. 89, 2), 
while Livy (Orosius) runs them up to " non minus quam triginta 
milia peditum." It is clear then that Livy (Orosius) followed some 
account of the battle which was more favorable to Pompey and 
less favorable to Caesar than Caesar himself. 1 

Plutarch (Pomp. 6g,foi., Caes. 42,7?^.) and Appian (B. C. II 70), 
who are generally believed to represent, at least ultimately, 
Asinius Pollio, agree perfectly with Caesar in his estimate of the 
forces engaged. Appian states that among many conflicting 
estimates he follows Roman authorities as the more trustworthy. 
The Roman authorities upon the battle were Caesar and his friend 
but faithful critic Pollio, and we have no account of the civil war 
emanating from the opposite side. A lost cause does not incite 
so many historians as a winning one. This Arnold of Rugby 
laments in his History of the Roman Commonwealth (Am. ed. 
p. 269), as soon as Caesar's Commentaries on the Civil War 
become his main authority. " The English reader," he says, 
" will, perhaps, have a more lively sense of its incompetence, if he 
considers what sort of a history could be drawn up of the events 
of more modern wars, if we had no other materials than the 
gazettes or bulletins of one party only." But some anti-Caesarian 
account of Pompey's last campaigns must have been accessible to 
Livy, and to this Pompeian version of the battle of Palaepharsalus 
we may fairly suppose that Livy went for items most favorable to 
Pompey, in whose cause he was such an enthusiast as to win the 
epithet of Pompeian from Augustus. And it may well have been 
this or a similar strictly partisan account of the battle, written in 
Greek and by a Greek, 2 with which Appian contrasts his Roman 
authorities with great parade of critical suspense. 

In Livy, then, who followed an account of the battle which 
certainly was not from so competent a witness as Caesar or Pollio, 
there may have been expressions of local description which led 
Frontinus to call the stream covering Pompey's right the Enipeus, 
and to say of Caesar's approach and order of battle, sinistrum 

1 Cf. Hugo Grohs, Der Werth des Geschichtswerkes des Cassius Dio (Berlin, 
1884). p. 69, where other proofs are given that Livy used some Pompeian ver- 
sion of the battle. 

2 This was very probably Theophanes Mytilenaeus. Cf. Grohs, /. c. p. 73. 


latus, ne circuiri posset, admovit paludibus. This, were there no 
indications of any kind to the contrary, might be taken as estab- 
lishing the fact that the Enipeus was the stream which covered 
Pompey's right, especially as Lucan has (Phars. VII 224 ff.) : 

At juxta fluvios, et stagna undantis Enipei 
Cappadocum montana cohors, et largus habenae 
Ponticus ibat eques. Sicci sed plurima campi 
Tetrarchae, regesque tenent, etc. 

Plutarch also (Brut. 4, 6) speaks of Pompey's camp on the day 
before the battle as n-pos e\ci>cWt ^copt'cur, and of Brutus escaping from 
the camp after the battle by a gate leading npos rfmov <fXa>8rj *a\ 
fiforov vbarwv km xaXa/xou. But with Caesar's deliberate expression 
against this identification of the Enipeus, and with the general 
arguments on military grounds against it, we must either deny the 
sufficiency of the authority for any contrary view, or must explain 
these passages otherwise. I should be content to balance the 
authority of Caesar, supported by the general military arguments, 
over against the unknown Pompeian source of Livy, supported by 
.rather vague concordances in Plutarch and Lucan, and choose the 
former. But another explanation of the language of Frontinus is 
possible. It may not rest on any statements of Livy, but be his 
own expansion and elucidation of Caesar's " rivus quidam impeditis 
ripis." As such it would certainly favor the view that the stream was 
the Enipeus, but, taking into account the fact that almost any name 
would do as well for the object which Frontinus had in mind, viz. 
a description of the strategical disposition of the forces on both sides, 
and the fact that the Enipeus was the main river of the scene, so 
that it would naturally suggest itself to one indifferent about and 
ignorant of the exact geographical details, the evidence is not 
strong enough to prevail against that on the other side. 

There remain to be considered only the statements of Appian 
concerning Pompey (B. C. II 65, 75) : Kal an-eo-rpaTcmeSevtre t» 

Kaicrapi 7rept &ap<ra\ov > Kal rpiaKovra o~Ta&iovt dWrjXov an€t^ov, and irapi- 
racrcre rovt Xoktovs it to p.era|v &apcra.\ov tc n6\cat Kal 'Eviireat lrorapov, 

ev6a /cm 6 Kalcrap ai>TicWdoy«i. The first statement puts a distance 
between the two camps which can harmonize with either Goler's 
or Napier's view of the position of Caesar's camp, owing to the 
width of even the right half of the valley of the Enipeus. The 
second statement is the sole ancient authority for locating the battle 
on the left bank of the Enipeus. It has led to impossible views. 


Mommsen assumes that Appian means Neopharsalus in distinction 
from Palaepharsalus. But in view of the other evidence it is not 
improbable that Appian purposely used Pharsalus loosely for 
Palaepharsalus, and it is quite possible that he blundered and 
failed to distinguish between the two. That this last supposition 
is not too harsh in the case of Appian can be shown from many 
worse mistakes. One example I have given in this Journal, Vol. 
V, p. 325 ff. As to the minor question whether the stream covering 
Pompey's right was the Enipeus, Appian's words do not necessarily 
imply this. They apply equally well to a line of battle parallel to 
the river. 

In conclusion, I consider it certain that both camps were on the 
right of the Enipeus, somewhere near the main route between 
Pharsalus and Larissa, and that the battle was fought at the base 
of the hills on whose slope Pompey's camp was pitched, near 
Palaepharsalus. I consider it probable that Palaepharsalus was 
on the hills north of the Enipeus and west of the main road 
north and south, that Pompey's line of battle extended east and 
west, parallel with the Enipeus, and covered on the right by a 
small stream running from the hills into the main river. This also 
makes Goler's position for Caesar's camp the more probable one. 
All these probabilities could be tested by explorations and exca- 
vations in the territory under consideration, similar to those carried 
on for Napoleon in France on the presumable sites of Caesar's 
encampments and engineering works. Such investigations are 
suggested by Seldner in the paper referred to. Possibly some 
member of the American School at Athens may yet undertake 

B. Perrin.