A minor (but frustrating) error in a Cretan gazetteer

“They said they were hair-dressers but … kind of hard to believe.”

Play It Again Sam

Woody Allen

I’m currently taking Mycenaean and Minoan sites from Crete and putting them into the Mycenaean Atlas Project database.  Today I was spotting sites in the very valuable gazetteer article:  Goodison and Guarita [2005].  Everything was going fine until I reached this:

“31. KHRONI KALYVI/ GANGALES

…

Location: After turning off the Ayii  Deka-Heraklion road towards Gangales,  140m on (just after the river) pass tiny chapel on left of road. 30m further on, cut across fields to the right (180m). The tomb is a few metres NW of a barn.” (1)

This tomb, possibly Early Minoan, is a circular tomb and mound of about 11 m. in diameter.  It was robbed in antiquity and, I suspect, of no really great importance or significance in itself beyond helping to fill out the picture of the extent of circular tombs in the Mesara during the EM.  Therefore the most important thing about it would be its exact position and that's what I want to establish.  There are a number of ambiguities in the directions to this tomb so I decided to take this down to the ground and provide a full description of what our authors are trying to tell us.

The main north-south road through the center of Crete runs from Heraklion to a junction with a major east-west road and very near to the town of Aghii Deka. (just a few km. further on one comes to Gortyn.)   And what our authors are saying is that if one turns towards Gangales (away from Ayii Deka to the east) then, after 140 m. one comes to a small chapel and then a road to the right which will lead to this tomb.

I tried every possible way to move 140 m. to the E from the N-S road to the main E-W road but nothing worked.  Actually 140 m. hardly gets you out of Ayii Deka.  There is a river in Ayii Deka and everything almost fit but didn’t quite.  Especially vexing was my inability to spot any tiny chapel on the edge of Ayii Deka. 

I spent a couple of hours on this and finally I just tried a guess.  Perhaps 140 was a mistype for 1400 (or something).  I started again tracing a 1400 m. route from the easiest turn-off from the N-S road to the east (another guess).  At the end of the 1400 m. trace I discovered everything together, the river, the chapel, the barn … everything.   And the ‘tiny chapel’ really was very tiny indeed; in Greece such a thing is known as a kandylakia – a roadside shrine/contribution box shaped like a full-size church.  And standing next to it was another even smaller and even more elaborate kandylakia.   The people of Greece sometimes erect these kandylakia at a place where a traffic death has occurred.  They form a kind of small refuge where a person might stop and pray during the day or light a candle or give a donation.

I don’t know what these kandylakia are called in Crete but ours look like this:

 

Chapel on L, kandylakia on R.

What about the route itself?

In this aerial photo the 1400 m. route is shown in light brown.  The river which Goodison & Guarita mention is in blue.  

Since our authors don't precisely say here is a close-up of the actual exit that Goodison et al. are referring to when they reference the start of this 1400 m. route.  Here's an aerial photo of it:

The route marked in red is the main highway from Herakleion to Ayii Deka.   The outskirts of Ayii Deka are shown on the left (W).  The turn that you make is labelled 'To the Tomb'.

If you're driving towards Ayii Deka then that turn is shown in the next photo from Google Street View.

Time for a close-up of the route's end and a tentative location for the tomb being sought.

This photo shows the route that you're driving on in light brown.  The river that you pass is in blue and the arrow A marks the position of the 'tiny chapel' and the kandylakia.  The line in purple is the 180 m. path that you take 'to the right'. 

Here is a close-up of the east end of the route.  The blue line is the river that Goodison et al. call out.  The lat/lon pair is the position of the 'tiny' chapel.  The circle is centered by the kandylakia and is 30 m. in radius.  Not surprisingly that brings us directly to the road 'on the right' which leads south 180 m. to the tomb.  Our authors say: 'Present structure: Barely discernible raised irregular circular mound of stones, overgrown by vegetation.   D: ?c. 11 m.  ...' (2)

In this picture there is some ambiguity because the tomb cannot now be seen.  I supposed that maybe the structure was to the SE and there is a circular feature there.  But it is not large enough.  Its diameter is only about 2.7 m. and we're looking for a feature with a diameter of 11 m. (approx.).  I also want to point out the feature that I've called a 'Burn Circle' at the lower left.  I often see these more or less circular features in the olive orchards of Greece; they are really just circles of ash where the farmers have burnt orchard rubbish.  Wishful thinking often converts these things into tholoi.

Goodison et al. would have done their research before 2005.  I was able to find the 11/16/03 image of this area in Google Earth.  It does appear to show an anomaly about 7.5 m. to the NW of the NW corner of the barn.  No other image that I examined shows anything at all in that position.  I have drawn a circle of 11.0 meters in diameter centered on that spot to suggest where the tomb may have been.  I emphasize that this is supposition based on what our authors have told us.

So, perhaps the best position for our sought-after tomb is  35.056539° N, 24.982953° E.


No one claims that their work is error-free.  Substituting 140 for 1400 is a tiny error in a really fine and useful article.   In the field of computer science this would be labelled as just a bug.  I shudder to think how many of these things there must be in my own work.  It does illustrate, however, one of the real advantages that digital representation has over traditional print venues for scholarly work.  This article is going to be around for a long time and the description for Tomb no. 31 will be misleading people for as long as it exists.   If the information in Goodison et al. [2005]  (and several hundred others that I could name) were in the form of a database then such a flaw could be repaired for free and turned around in seconds.  It's just like software in that respect.  No professional expects software to be completely free of errors.  No moral opprobrium attaches to this.  Software authors recognize better than most other professionals that, when it comes to details, the human mind is basically trash.  As a result a large part of software implementation consists of making provision for keeping flaws under control.

One of those devices is the version numbering system.  As hideous as it may sound to an academic it may be time to start versioning articles and books.  The versioning information goes along with the article/book and would explain what has changed since the last version and why.  Of course this can only happen if these articles/books are online.   A consummation devoutly to be wished.

Of course I also think that articles and books shouldn't be printed at all.  It's outmoded.  Why commit errors to epistemological prison?

Finally, let me express my gratitude to Goodison and Guarita for providing this gazetteer from which I've learned so much.

~~~~~~~~~~~~~~~~~~~~~~

A new database for the Mycenaean Atlas Project (www.helladic.info) was delivered on May 11, 2018.  Approximately 35 new sites were added in the Siteia region of Crete.  The new DB is rev 0059.

Also don't miss my review of Against the Grain by James C. Scott.  It's here: http://mycenaeanatlasproject.blogspot.com/2018/02/some-notes-on-james-c-scott-against.html

If you like these posts then please follow me on Twitter or Google Plus (Robert Consoli).

You can e-mail me (and I hope you will) at  bobconsoli   at   gmail.com

And please remember - Friends don't let friends use Facebook.


If you'd like to have a copy of the Mycenaean Atlas database then e-mail me and tell me about your project.  And remember that useful .kml and/or .csv files can be generated directly from all the windows of the website helladic.info   Try it out!

Notes

(1) Goodison and Guarita [2005],  186.
(2) idem.

Bibliography

Goodison and Guarita [2005]:   Goodison, Lucy and Carlos Guarita, 'A New Catalogue of the Mesara-Type Tombs', Studi Micenei ed Egeo-Anatolici [47], pp. 171-212, 2005.

Using the Mycenaean Atlas Project – 3

The Find Type table

In previous posts I have discussed the ‘site’ table, the ‘fnb’ (site bibliography), and the‘biblio’ (full bibliography) tables.

 In this post I want to discuss the find type table (‘findt’).  The Mycenaean Atlas Project is a database that is intended to support specific lat/lon pairs for each Mycenaean site.    In addition to this primary purpose users of this DB should be given some idea about the nature of the site and when it was created/occupied.   The find type (findt) table contains a minimum of information about the nature of the site and in such a way that it allows useful join queries.  The fields of the ‘findt’ table are as follows:

a.     Place key (pk):  This external key is identical to the primary key, pk, in the ‘site’ table and the foreign key, pk, in the fnb table.  It consists of the letter ‘C’ followed by an integer.  E.g., ‘C100’.   This allows join queries to be formed such as:

select * from site a, findt b where a.pk = b.pk;  // this will return the combined info for site and site type from the database.

b.     Source (src): This is the bibliographic source symbol for the source which gives the find type for this site.  It is the same as the src field in the ‘fnb’ and the ‘biblio’ tables.  This allows useful join queries.

c.      Page (Page): This is the page in the bibliographic source which gives the find type for this site.

d.     Type (Type): This is a string of up to twenty characters that gives the actual type of this site.

e.     Comment (Comment): This is a string of up to 1024 characters that characterizes the site in more detail and usually as a quotation from the source in the Src field.

Here’s an example of the full 'findt' record for tholos tomb C237 (Koryphasion):

Query: select * from findt where pk = ‘C237’;

C237, S81, 116, Tholos, 'The tholos tomb excavated here is the earliest known in Messenia and may indeed be the earliest tholos tomb on the Greek mainland. '

C237, B, 405, Tholos, ... medium sized tholos tomb, constructed In late MH times. Variously referred to as Korifásio, Osmánaga or Haratsári

C237, MS61, 242, Tholos, '..this is the earliest known mainland tomb of tholos type (in Messenia at any rate)..'

C237, Z, 451, Tholos, '...eines der ältesten Tholosgräber des griechischen Festlandes. '

C237, GAZ, 1, Tholos, '...the early date of the finds that indicate it is one of-if not the-earliest tholos tombs on mainland Greece ... '

C237, A62, 86, Tholos, 'Kourouniotis grub hier 1926 ein Tholosgrab aus.'

C237, OP, 198, Tomb, 'La tombe est qualifiée par Kourouniotis de "souterraine".'

Each record consists of the pk ID: here 'C237', the symbol for the supporting work, the page number in that work, the type: Tholos or 'Tomb', and a comment from the source about this site.  

The supporting sources are returned in symbolic form (e.g., 'S81').  Their short citation forms are:

Ålin [1962],  Boyd [1999],  Heath [nd]   (the PRAP on-line gazetteer),  McDonald and Simpson [1961],  Pelon [1976],  Simpson [1981],  Zavadil [2012]

So, the find type of C237 (Koriphasion) is ‘tholos’ and this is backed up, in this case, by seven source citations.

And whenever you request the lat/lon fields in your query you can turn the results directly into a map.

So, just for fun, where are all the sites in the Atlas that are typed as 'Cist Tomb' only?

Cist Tombs only.  The query was:
select a.pk, a.lat, a.lon from site a, findt b where a.pk = b.pk and b.type = 'Cist Tomb';

Useful Sample Queries using the findt table

select * from findt where pk = ‘C100’;  // returns all the find type supporting records for site 'C100'.

select count(*), type from findt group by type;   // returns just the distinct types and their counts.

select pk from findt where type = 'Tholos' or type = 'Cem' or type = 'Chamber Tomb' or type = 'Mound' or type = 'Tomb' or type = 'Tumulus' or type = 'Cist Tomb' or type = 'Grave';

  // returns the place key and name for all funerary type sites from 'findt' and 'site'.

select pk from findt where src = ‘OP’;  // this returns the IDs for all sites for which Oscar Pelon was used as a source.

select distinct pk from findt where src in (‘S81’, ‘S14’);  //returns those site ids documented by Richard Hope Simpson.

select a.shortc, a.longc from biblio a where a.src in (select distinct b.src from findt b where b.pk = 'C237');   // this join returns short and long citation forms for all the sources used to document site ‘C237’.

This diagram shows the relationship between the several tables discussed so far.

In my next post I'll discuss relative vs. absolute chronology in the Mycenaean Atlas Project along with the use cases that the database must support.

Anyone who would like to have a copy of the MAP database can send an e-mail to bobconsoli 'at' gmail.com or leave a comment on any of my posts.  To run the MAP database requires a SQL server running on your desktop computer.   MySQL is such a server and it is powerful, industry-standard, and free.  

I can and will make .kml or .kmz files, which can be opened directly in Google Earth, available to those who would like them.  

I can also create .csv files for people who would like to import Mycenaean Atlas Project data into Google Earth but would like it in tabular form.

Those who do not have a SQL server but would like the full database in .pdf form can have that for the asking.

If you like these posts then please follow me on Twitter (Squinchpix) or on Google+   (Robert Consoli)

Facebook?  Sorry.I.just.can't.

Bibliography

Ålin [1962]:  Ålin, Per, Das Ende der Mykenischen Fundstätten auf dem Griechischen Festland. Carl Bloms Boktryckeri A.-B., Lund, 1962.

Boyd [1999]:  Boyd, Michael John.  Middle Helladic and Early Mycenaean Mortuary Customs in the Southern and Western Peloponnese.  University of Edinborough.  Scotland.  1999.

Heath [nd]:  Heath, Sebastian.  "PRAP Site Gazetteer" http://classics.uc.edu/prap/static/sites_list.xsl.html

McDonald and Simpson [1961]:  McDonald, William A. and Richard Hope Simpson. 1961. Prehistoric Habitation in Southwestern Peloponnese.  American Journal of Archaeology. Vol. 65, No. 3 (Jul., 1961), pp. 221-260.

Pelon [1976]:  Pelon, Olivier.  Tholoi, tumuli et cercles funéraires; Recherches sur les monuments funéraires de plan circulaire dan l'Égée de l'Âge du Bronze (IIIe at IIe millénaires av. J.-C).  Bibliothèques de l'École française d'Athènes et de Rome - Série Athènes, 229. 1976.

Simpson [1981]: Simpson, Richard Hope. Mycenaean Greece. Park Ridge, New Jersey: Noyes Press, 1981.

Zavadil [2012]: Zavadil, Michaela. Monumenta: Studien zu mittel- und späthelladischen Gräbern in Messenien.  Wien:Osterreichische Akademie der Wissenschaften Philosophisch-Historische Klasse Denkschriften. 2012.