The ‘Cyclopean Wall’ at the Isthmus: Some Observations

 

“The plays of William Shakespeare were written by William Shakespeare or, perhaps, by someone else who had the same name.”    

Anonymous

In 1954 Dr. Chrysoula Kardara was working on the top of Rache (‘the Ridge’) which is in the area of Isthmia just south of the well-known sanctuary of Poseidon.  She happened to look east across the valley separating Rache from the Mytika plateau and saw (some 220 m distant) a stretch of walling that appeared to be of Cyclopean work.  Closer inspection showed that it was, indeed, very probably Cyclopean.   Further investigation by Kardara and Dr. Oscar Broneer turned up further segments that appeared to be of the same or similar construction and which could, plausibly, be part of the same structure.  Kardara and Broneer first believed that these segments could be part of the support substructure for a road.[1]   In the ensuing ten years, however, Broneer began to suppose that these segments were part of a great defense wall.  His arguments were first elaborated in an article from 1966 in which he suggested that not only were all these segments a defense wall but that what was being specifically defended against was the Dorian Invasion; as he put it: ‘The Return of the Heracleidae’.[2]  Nor was he alone; Several major scholars were convinced by Broneer’s change of mind.  Richard Hope Simpson enthusiastically defended this thesis in his work with Hagel on Mycenaean roads and bridges [3].  I shall have more to say about Dr. Simpson’s defense of this thesis. [4]

Another defender of Broneer’s thesis was the scholar and archaeologist, James Wiseman.[5]  Winter [1971] also adopted Broneer’s thesis, having no difficulty accepting the shallow buttresses (e.g. in section Pe) as towers. [6]

As time passed, however, other archaeologists began to express doubts.  The discoverer of the first segment, Dr. Kardara, as well as the younger scholars, Morgan, Gregory, Loader, etc. weren't so easily convinced of the unity of this 'wall'. 

Catherine Morgan provides a careful discussion of all segments of the wall.  She starts by saying "There is no evidence to indicate that it crossed, or was intended to cross, the Isthmus, or that it continued to Corinth and served to connect rural settlements; ... "  Loader  shows that some of the segments have widely varying dates.  Timothy Gregory  says " ... there is no evidence to connect these short sections into a great defensive work across the Isthmus; some of the sections might not be Mycenaean at all." [7]

How to defend an Isthmus

It is natural for armchair strategists to look at a map of the Isthmus and suggest a defense of the Peloponnese based on a trans-Isthmus wall. Professional strategists, on the other hand, will look at the same map and see a very different defense; one cheaper, quicker, and more effective. The Isthmus is almost entirely blocked off by Mt. Geraneia which forms an admirable defensive line all by itself.   What is required is to block off the Scyronian road on Geraneia's east, secure the pass over the center of the mountain and block the narrow shore along the Corinthian Gulf on the west.  [8].  All of these measures are well within the ability of even a modest force.

If this suggested line fails then another line can be formed by starting at Loutra Helenis [C7732] on the Saronic, continuing across Mt. Onium, and ending at the Acrocorinth. [9]  But, as Herodotus saw, neither wall nor defense line of any type will work if the enemy has access to boats.  If invaders have a navy, then the Peloponnese is not defensible by anything except another navy. [10]

Despite all this, there have been wall-building attempts at or near the Isthmus. But, since the BA, we know of two walls that actually crossed the Isthmus.  The first is the defense wall of the Peloponnese which was built in 479 BC.  The other is the Hexamilion wall which was begun in the fourth century A.D.  We know that it crossed the Isthmus because its entire route may be easily traced today. [11]

In 2022-4 I undertook, with the help of associates, to find out where the several fragments of the ‘Cyclopean Wall’ were actually located and to photograph them if that were possible.  The more or less unspoken goal of this activity was to try to understand the state of the question with respect to a Bronze Age trans-Isthmian wall.

These associates were able to find a number of the named segments and I have written about the most significant elsewhere in this blog. [12]

Our quantitative findings are given in the following table.  The column headed ‘M.A.P.’ refers to the Mycenaean Atlas Project.  Labels in the form ‘C____’ refer to actual BA sites, here the several segments.  Labels in the form ‘F____’ refer to features which are either modern (churches, towns, etc.) or to ancient sites which cannot be referred to the Bronze Age.  Both of these label forms can be directly searched for in the M.A.P.

Name	Length m.	Width m.[13]	Lat	Lon	Elev m.	Above Ground m.	M.A.P.
Sk	17	4.0-3.75	37.913134°	23.006914°	5-7	0	C7758
St	25	4.5-5.75	37.913405°	23.004475°	13	0.5	C7759
Ro	8.4	?	37.913475°	23.003006°	27	0	C7760
Ph	11.7	?	37.913669°	23.000387°	55?	?	F7690
Pe	45.5	3.60-4.0	37.913457°	22.997472°	65	0.5-1.0	C7761
Sp	22	?	37.911523°	22.994357°	67	2.5	C7769
Zo	13	?	37.91063°	22.992883°	74	?	C7765
Vl	30	?	37.908761°	22.989498°	?	?	F7692
Pa	12.5	?	37.908077°	22.988464°	107	‘2 courses high’	C7762
Mi	4.5	?	37.914877°	22.993647°	60	0.5	C7763
Ge	20.5	3.4-3.6	37.915184°	22.993201°	56	0.5	CC7764

This second table offers some auxiliary information about each segment …

 

Name	Face	‘Tower’	Flat ground or Slope?
Sk	N/S	n	Flat
St	N/S	‘setback’	Flat
Ro	N	y	Slope
Ph	N/?	y	?
Pe	N	y	Slope
Sp	N	n	Slope
Zo	S	n	?
Vl	N	n	Slope
Pa	N	n	Slope
Mi	?	n	Flat
Ge	N/S	n	Flat

 

This material can be made clearer with a few pictures.  First a picture of this area from the north-east.  The Corinth Canal is out of the picture on the bottom.  At the top edge is the range of the Oneion hills.  The Mytika plateau is in the center; the ridge of the Rache is to the right (W) of it and lit up by the morning sun.  The various discovered segments of wall are outlined in green.  They are proportionately sized to their length at ground level.   The vertical is exaggerated for visibility; most of the segments are level with the ground.  Pa (furthest away) is at the head of the valley between Mytika and Rache.  Its position is known to within a few tens of meters.

As can be seen the ‘trans-Isthmian’ wall consists of a number of widely scattered segments.

The next picture shows the same scene looking directly south and from a lower angle.  This makes it possible to appreciate the relative elevations of the several segments.

The next picture attempts to join the several segments into what I consider to be the most probable groupings.

It seems likely, for example, that Sk/St/Ro form a single stretch and one motivated by a single intention.  Pe/Sp seem, likewise, to have been similarly motivated as a single construction.  On the west end it appears that Mi and Ge are overwhelmingly likely to have constituted an intended single entity.  The difficulty is, of course, the stretch from Zo to Pa.  The little gully between Sp and Zo has been the scene of many modifications over the millennia and the remains appear hopelessly confused.  Between Zo and Pa there is an additional segment of walling called ‘Vl’ (not shown here).  Because no excavation or other work has been performed on this steep hillside it is pointless to speculate about how, exactly, Zo is connected to Pa or whether builders ever conceived them to be part of the same structure.  

Here is a variation on the previous picture which concentrates on the ‘voids‘ or missing segments between the several groupings.

Here the missing stretches are shown as orange lines.  The two largest are, of course, that between Ro and Pe on the one hand, and Pa and Mi/Ge on the other. 

Pa to Mi

Let us deal with the latter first.  The stretch between Pa and Mi would run about 900 meters at the slope break of the Rache.  There is no evidence of any Cyclopean work here.[14]  The usual suspects are blamed, WW II gun emplacements, agricultural activity, erosion, etc.[15]

It is very tempting to conclude that there never was a stretch of walling that joined Pa to Mi.  This is either because Pa and Mi were constructed for intentionally different purposes (and at different times and by different actors) or because Pa and Mi were constructed for a related purpose but that that specific purpose was satisfied by the steep slope of the Rache without any additional construction.  How can that be?

The builder/designers among the Mycenaean people often showed a surprising sensitivity  to the actual configuration of the landforms among which they worked.  It seems that at least some of them considered the various hills of their homeland as a kind of raw material to be utilized as it suited their purposes.  The best example of this was the terra-forming (there is no other word) that was performed on the northern border of Lake Copais in Boeotia.  Lake Copais was an actual lake at the time the earliest Greek speakers entered the valley.  Several important streams entered this valley but, as this was a karstic landscape, no outlet could be found for these waters until they reached the level of the several katavothrai or sink holes which would direct the river flow out of Kopais and towards the ocean.  This occurred when the water reached about 97 m. a.s.l.  The lake had a permanent typical depth of three or four meters.  Two of the significant streams that flowed into the Copais Basin are the Cephissus and the Melas.   Before any land reclamation was done the Melas had already carved a channel to the nearest Katavothra some 17 km to the E.  The Cephissus, on the other hand, flowed directly into the lake basin.  At some point it appears that the Mycenaeans of Orchomenos determined upon a scheme to drain the Lake by directing the waters of the Cephissus into the channel of the Melas.  A secondary channel was created for this purpose.  The combined flow of the rivers, however, was too great for the original channel of the Melas and it necessarily overflowed.  So, it appears that the second step in the Minyan plan was to create a great dike, south of the channel of the Melas, which extended all the way to the katavothrai.  When the combined flow of the two rivers was too great for the original channel then the overflow would be contained by the dike.   As a result the Copais lake began to drain.  The point in all of this is that no dike was constructed north of the channel of the Melas.  The Minyans relied on the steeply plunging hills on the north to contain any overflow in that direction. [16]

Another striking example of the sensitivity of the Mycenaean builders to landscape forms is the massive land movement required in order to create the new harbor at Pylos in Messenia.

Ro to Pe

The fact that no traces of the BA wall have been located between Ro and Pe is often explained by the fact that the Hexamilion Wall also was built in that location and, presumably, destroyed traces of the earlier hypothetical BA wall.  The distance (length of the void) between Ro and Pe is 470 m.  The Hexamilion Wall parallels/occludes the (hypothetical) BA wall for a distance of 240 m or only about half the length of the void.   There have been sporadic and unexcavated finds between the Hexamilion and Pe along the supposed line of the BA wall, but we are not aware of any finds that would establish a continuity in this area.[17]  One of the hypotheses we are forced to entertain, then, is that the ‘intention’ that motivated Sk/St/Ro differed from that that motivated Pe/Sp.

What hypotheses are we left with?

           

1     These wall fragments are parts of a gigantic fortified cross-Isthmus wall meant to defend the Peloponnese from an invasion.

2     These wall fragments are parts of a single-purpose cross-Isthmus wall.

3     These wall fragments are all parts of a single-purpose local wall encompassing Sk to Ge.

4     These wall fragments may be grouped into several components; each one of which was constructed with differing intentions.

a.    Pe-Sp-Pa-Mi/Ge were constructed with a single intention relating only to this valley; e.g. a hunting kite or to support a road.

b.    Pe-Sp-Pa were constructed with the purpose of buttressing a road.

c.    Sk-St-Ro were constructed with the intention of defending only the coast from the N.

d.    Mi-Ge were constructed with some purpose relating only to the area around the later Poseidon Temple.

5     These wall fragments were each constructed with differing intentions.  That is, no segment is related to any other.

We are now in a position to suggest that some of these hypotheses appear less credible than others.  If I had to rank them in order from the most probable to the least probable I would say this: {4b and 4c, 4d, 4a}, 5, 3, 2, 1.  Here the bracketed alternatives seem, to me, about equally probable.

In eighteen months this is what my associates and I have been able to learn about the debate over the wall and the remainders of it.

I wish to end this post with a heart-felt and sincere thank-you to my associates Peter Barkevics and Hajo Becker for their hours of field work finding and photographing the remains of the 'Cyclopean Trans-Isthmian Wall'.

Footnotes

[1] Kardara [1971] 85.  Broneer [1966] 351, fn. 9: “This part of the wall was first observed in 1954 by Chrysoula Kardara, who was then excavating on the Rachi on the opposite side of the gully. We were then under the impression that it was part of a retaining wall for a road. See Hesperia, XXIV, 1955, p. 124”  Also Broneer [1955] 124. Hesperia, XXIV, 1955, p. 124: “It is clearly a retaining wall with a finished face toward the gully. No foundations of other buildings are visible in the vicinity and almost no pot sherds were found, either in the fill in front of the wall or in the near-by plowed field. The wall can have been used only as a retaining wall for a road, probably a part of the ancient communication system connecting the Isthmia region with Epidauros and Troizen.”

[2] “Return of the Herakleidai” in Broneer [1966] 357: “ … the influx of new tribes and the decimation and displacement of the Achaean inhabitants.  These movements, known in mythology as the Return of the Herakleidai and in history as the Dorian Invasion.”

[3] Simpson and Hagel [2006]  124-43.  On 124: “These, together with historical factors and common-sense observations, all show that Broneer’s main conclusion is substantially correct, i.e. that the Wall was indeed intended as a fortified barrier and boundary demarcation across the Isthmus.”  Emphasis added.

[4] Simpson and Hagel [2006] 124, fn. 61 : “In Isthmia VIII the Wall is usually described as “Mycenaean Wall” or “Wall,” in quotation marks, which may be intended as a neutral qualification, but which in fact convey an impression of disbelief.” Emphasis in the original. The work Simpson is referring to is Morgan [1999] and here he  clearly emphasizes ‘belief’ as the desired approach.

[5] See Wiseman [1965] 18 ff.  “The Dorians could have had, though, a somewhat greater hand in events than Vermeule and others are willing to allow. There is, after all, a Mycenaean fortification wall that was built across the Isthmus of Corinth about the middle of the LH IIIB period. … There are towers projecting from the wall towards the north and east indicating that it was manned by Peloponnesians.”   Here Wiseman provides a classic example of the fallacy called ‘begging the question’.   It is exactly the existence of the ‘Mycenaean fortification wall’ that needs to be proved.

Also Wiseman [1978] 59-60.  He refers uncritically to the “Mycenaean Wall” throughout.  Also 59 “Pottery found between the two faces of the wall and on the inside of the faces in the packing of the interstices date to the end of L.H. IIIB and the beginning of L.H. IIIC.  The pottery therefore indicates a date of construction not far from 1200 B.C.” And 60 “The features of the wall pointed up by the question are suitable for a fortification wall, not for a road.” 

[6] Winter [1971] 152 and esp. fn. 5.

[7] Morgan [1999]; Loader [1995] 164-7; Gregory [1993] 4.

[8] Pass over Mt. Geraneia : Mycenaean Atlas  Project F7678.

[9] James Wiseman lists defenses of the Onium line at Wiseman [1978] 77, fn. 92.

[10] Herod. ix.9.2.  At this point (sometime in 479 BC) the trans-isthmian wall has been completed but it looks, momentarily, as though the Athenians will come to an agreement with the Persians.  If that happens, Chileos tells the Spartans, then no matter how strong a wall is flung across the Isthmus its gates will be opened to the Persians since they will have the Athenian navy as a support.

[11] For the Hexamilion wall: Gregory [1993].

[12] For our activity in general see this. For segment Pe see this and our relocation of it here.  Ro is discussed here, Sp here, and St here.  Mi and Ge are part of the discussion here.  Pa is part of the discussion here.  When I wrote this post we had still not relocated either St or Ro and, as a result, all of its speculations are moot.

[13]  Loader [1995] 166, 9.1.5.ii, ‘Average Dimensions’, “As the function of the Isthmian wall is presently unknown, and the remains are suggested by this author to represent sections of more than one structure, comment on its dimensions must be limited.”

[14] It is often said by the unthinking that ‘Absence of evidence is not evidence of absence’.  But it is consistent with it.

[15] Simpson and Hagel [2006]:  127. “Between Section Pa and Section Ge the Wall must have descended gradually along the southeast slope of the Rachi ridge to the Sanctuary area.  Most of this part is now lost, largely due to later activity here; … ”  Morgan [1999] 444 refers to one section of Pa as closed off or blocked by a WW II military trench.  Simpson says ‘Most of this part is now lost … ‘ but, in truth, beginning with the two small sections of Pa at the head of the ravine, there is no further trace of a Cyclopean wall until Mi is reached, some 900 m. to the N.

[16] A more complex story than I have sketched here.  Much is suppositious.  We can only guess the intentions of the Minyans from what we know of the actual dikes that survive.  See Simpson [1980] 60, fig. 6 and 61, ‘C1. Ancient Orchomenos’.  And many publications by Knauss such as Knauss et al. [1984].  Interested readers can find a list of our blog posts on this topic here.

[17] Trenches MW 3 and MW 4.  See Morgan [1999].  For MW 3 see 442, ‘4. Trench MW 3’.  For MW 4  see 442, ‘6. Trench MW 4’.

Bibliography

 

Broneer [1955] : Broneer, Oscar. ‘Excavations at Isthmia, 1954’,  Hesperia: The Journal of the American School of Classical Studies at Athens (24) 110-141.  1955.  Online <a href=&quot;https://www.jstor.org/stable/147264&quot;>here</a>.

Broneer [1958] : Broneer O. The Corinthian Isthmus and the Isthmian Sanctuary.   Antiquity. 32(126):80-88.  1958.  Online <a href=” https://www.cambridge.org/core/journals/antiquity/article/abs/corinthian-isthmus-and-the-isthmian-sanctuary/B278886913654FB118B7D316DC321596”>here</a>.

Broneer [1959] : Broneer, Oscar. 'Excavations at Isthmia; Fourth Campaign, 1957-1958',  Hesperia (28), 334.  1959.  Online <a href="https://www.jstor.org/stable/147250">here</a>.   Especially p. 134, 'Pottery', nos. 1-3 and fn. 43 which mentions pottery finds ‘in the village of Isthmia’ which refers, presumably to section St.

Broneer [1966] :  Broneer, Oscar. ‘The Cyclopean Wall on the Isthmus of Corinth and Its Bearing on Late Bronze Age Chronology’, Hesperia: The Journal of the American School of Classical Studies at Athens (35:4). 1966.  Online here.

Broneer [1968] : Broneer, Oscar. 'The Cyclopean Wall on the Isthmus of Corinth, Addendum', Hesperia: The Journal of the American School of Classical Studies at Athens (37:1), 25-35.  The American School of Classical Studies at Athens. 1968.

Gregory [1993] : Gregory, Timothy E., Isthmia V. The Hexamilion and the Fortress, American School of Classical Studies at Athens. Princeton, New Jersey. 1993. ISBN: 0-87661-935-9.

Kardara [1971] : Kardara, Chrysoula. 'The Isthmian Wall; (A Retaining Wall for a Road)', Athens Annals of Archaeology (4:1), 85-89. 1971.

Knauss et al. [1984] :  Knauss, Jost with B. Heinrich and H. Kalcyk. Die Wasserbauten der Minyer in der Kopais – die älteste Flussregulierung Europas.  Technische Üniversitat München. 1984.  ISSN : 0172-7494.

Loader [1995] : Loader, Nancy. The definition of cyclopean: An investigation into the origins of the LH III fortifications on mainland Greece I, Durham theses, Durham University. 1995. Online here.

Morgan [1999] :  Morgan, Catherine. Isthmia VIII; The Late Bronze Age Settlement and Early Iron Age Sanctuary. The American School of Classical Studies at Athens, Princeton, New Jersey. 1999.

Simpson and Hagel [2006]:  Simpson, R. Hope and D.K. Hagel. Mycenaean Fortifications, Highways, Dams and Canals. SIMA vol. 133. Paul Åströms Förlag. Sävedalen, Sweden. 2006.

Winter [1971] : Winter, F.E., Greek Fortifications, University of Toronto Press, Toronto and Buffalo. 1971. ISBN : 0-8020-5225-8.

Wiseman [1965] : Wiseman, James.  "Greece and the Early Greeks", Arion: A Journal of Humanities and the Classics (4:4) 700-720. 1965.  Online <a href="https://www.jstor.org/stable/20162995?seq=1">here</a>.

Wiseman [1978] : Wiseman, James. The Land of the Ancient Corinthians. Paul Åströms Förlag, Göteborg, Sweden. 1978. ISBN: 91-85058-78-5.