Peter's Ancient History and Religion Pages

Peter's Index Peter's Index  Ancient History Home Ancient History Home  Samos and Athens Samos and Athens 



Inscription IG I3 11: Alliance of Athens and Egesta: 458/7 or 454/3 or 418/17 BCE?

Friday 14th May 2004
by Peter Eyland

Dating the Egesta Alliance

Note: February 2015 the greek has been updated, to be viewed correclty in all browsers and on iPads and iPhones.

Introduction

An alliance was made between Egesta (a town in NW Sicily) and Athens.
The evidence for this alliance comes from an inscription on two fragments of a marble stele, labelled variously Inscriptiones Graecae I2 19,20 and I3 11.
The alliance has been dated as 458/7 or 454/3 or 418/17 BCE (Fornara, 2003:81).
The salient features of the inscription are given.
Evidence for the first two dates is discussed and then the claims of letter-form analysis.
Evidence for the third date is discussed and the question of Thucydides' silence about the alliance is raised.
The consequence of the alliance for Sicilian politics is discussed.

The marble inscription

The top three lines of the stele (Wick, 1995:plate XXIII) are given below showing its 48 character stoichedon (grid) pattern and the worn character of the marble. A possible reconstruction from Meiggs and Lewis is shown as well. The name "Egesta" is in the larger letters at the end of the first line.

inscription
transcription

A section of line 14 from Wick (1995:plate XXIII) is shown below with two important letters. The three-barred sigma (looking like a backward Z) is second from the left and last on the right. The round tailed rho is third from the right (looking like an R).


line14

A section of line 3 from Wick (1995:plate XXIII) shows the Greek letters WN (Omega Nu) usually written ων (English: ōn) in the centre. There is a black mark on the lower left hand side of the W. The picture also shows the region of the two worn letters to the left of ων and on the right hand side of ων are the first two letters of ΕΡΧΕ (usually written ἔρχε ), with a round tailed rho.


line2

Dating the inscription directly depends on the identity of the Athenian archon in line 3. However due to abrasion of the marble, there has only been universal agreement on ων , the last two letters of the archon's name. Appendix 1 gives the names of Athenian Archons from 528 to 292 BC (Bickerman, Chronology:137,138) and those ending with ων (English: ōn) have been highlighted. Restricting possible archons to those from the start of the first Peloponnesian war, until the start of the final phase of the war, gives the five names in the second column, viz: Habron, Ariston, Epameinon, Aristion, and Antiphon (Ἅβρων , Ἀρίστων , Ἐπαμείνων , Ἀριστίων and Ἀντιϕῶν ).

Argument for the date 454/3 BCE

H.G. Lolling (1891) [1] was the first to place inscription IG I2 19 in the historical context of Diodorus' Egestan war.
Diodorus (Library, 12.86.2) wrote:

In Sicily a war arose between the peoples of Egesta and Lilybaeum over the land on the Mazarus River, and in a sharp battle which ensued both cities lost heavily but did not slacken their rivalry.

κατὰ δὲ τὴν Σικελίαν Ἐγεσταίοις καὶ Λιλυβαίταις ἐνέστη πόλεμος περὶ χώρας τῆς πρὸς τῷ Μαζάρῳ ποταμῷ· γενομένς δὲ μάχης ἰσχυρᾶς συνέβη πολλοὺς παῤ ἀμϕοτέροις ἀναιρεθῆναι καὶ τῆς ϕιλοτιμίας μη λῆξαι τὰς πόλεις.


Since Diodorus had dated this war to 454/3 BCE "while Aristōn was archon of Athens" ( ἐπ' ἄρχοντος δ'Ἀθήνησινʹ Ἀρίστωνος· Library, 12.86.1), Lolling also dated the inscriptional alliance to that year. Also de Gaertringen, (IG I2, 1974:11) printed Aristōn, as did M.N Tod [2].

In a search for historical context, Meiggs (1975:599) noted that around this time there was "buoyant mood" in Athens because of the reforms of Ephialtes [3] and the first successes in Egypt. Meiggs considered these factors to be consistent with dating the alliance at this time. Meiggs and Lewis (1969:81) note that 454/3 BCE was the "generally accepted" date for the alliance until sometime around 1969 [4].


Argument for the date 458/7 BCE

In 1944, evidence was given in favour of the archon Habrōn. Raubitschek [5] (1944) read the worn text as [br i.e. with a probable beta and rho. This, if accepted, would have ruled out Aristōn in favour of Habrōn, and fixed the alliance date a few years previously in 458/7 BCE.

In 1949 Klaffenbach [6] read ]ab i.e. a definite alpha in front of the probable beta, but Pritchett [7] (1955), Meritt [8] (1964), and Mattingly (1963:268) rejected this completely.

In 1964 Meritt read no rho, i.e. the rho was not readable, but Habrōn was left as the probable reading because of the probable beta. By 1971 Meritt (according to Wick, 1995:187,n.8) had moved to Raubitschek's position.

The move away from 454/3 BCE needed a revision of Diodorus' historical context. Meiggs and Lewis (1969:81) note that Diodorus' text was "unsatisfactory" [9]. One reason was that the Lilybaeum mentioned was not founded before the fourth century (Ehrenberg, 1975:456,n.138) and so there could not have been a war with that city in the mid fifth century. It was proposed that Λιλυβαίταις was a mistake for Ἀλικυαίοις .

Also Beloch [10] had proposed an emendment that added πρὸς Σελινουντίους to the original. This emended Diodorus' text (Library, 12.86.2) to be:

In Sicily a war arose between the peoples of Egesta and Halikyai against Selinus over the land...

κατὰ δὲ τὴν Σικελίαν Ἐγεσταίοις καὶ Ἁλικυαίοις ἐνέστη πόλεμος πρὸς Σελινουντίους περὶ χώρας


This moved its context to 416/415 BCE: the time when the Leontines and Egestans sent ambassadors to Athens asking for help against the Syracusans (Diodorus, Library: 12.82.3; Thucydides 6.6). The alliance inscription was then freed from Antiphōn's context. This still fitted Meiggs (1975:599) "buoyant mood" as it meant a shift of only four years.


Letter-form Evolution

Meiggs (1966:86) observed that inscribing the name of archons in the prescripts of decrees was customary from the Peace of Nicias in 421 BCE onwards [11]. Meiggs mentioned the alliance with Egesta as an example of this practice and so a date for the Egestan alliance after 421 BCE was indicated. Before looking at a date that is not in the 450s, an important argument must be considered. This relates to the shape of Greek alphabet letters and how they evolved over time.


Kavvadias [12] (1897) first drew attention to an earlier form of sigma (early sigma) on the Athena Nike inscription. Segre (1938:16) and Raubitschek (1940:477f) asserted that the three-barred form (three barred sigma) was replaced by the four-barred form (four barred sigma) on inscriptions after 445 BCE. On this basis, the dates of a number of inscriptions with three-barred sigmas (e.g. Egesta) were moved from the 420s to the 450s BCE.

Mattingly attacked this letter-form analysis and asserted that historians had been "imprisoned by a dogma" (1961:149) and an "over-rigid application of epigraphy" (Mattingly, 1961:173). Since then Mattingly has argued copiously [13], on the basis of historical congruence and textual context, for a number of inscriptions to be put back into the 420s BCE. For example, Mattingly argued that the Coinage Decree suited c.420 admirably, the Decree of Kleinias had an affinity with 426/5 and the Athena Nike temple was not built in the 440s etc.

Meritt and Wade-Gery agreed that letter-form dating was not self evident but "true in fact" (1962:67) and then argued that the historical context of these decrees pointed to earlier dates than Mattingly, in conformity with letter-form dating.

When Mattingly (1963) dated the Egesta alliance, he first looked to the 420s and suggested that Epameinon (429/8 BCE) or Aristion (421/0 BCE) might be appropriate (1963:268). However, he decided, from the inscription itself, that a dotted phi was appropriate in front of "ων" (ων) . This consequently led him to propose Antiphōn and 418/7 BCE for the alliance date (Mattingly, 1963:269).

Meiggs' particular contribution (1966:86-98) [14] was to use letter forms to "formulate criteria" and "provide the framework" for dating fifth century Attic inscriptions that could not be dated by other means. Meiggs first tabulated dated decrees (1966:92) and then undated decrees (1966:94).

Mattingly (1967:27) has conveniently summarised Meiggs' criteria as follows:

        were not used after 445 BCE
 was used dominantly from 448 BCE.

 was not found after 438 BCE    was not used after 430 BCE.

Mattingly (1967:27) in reply to Meiggs (1966), argued that "spelling, grammar, idiom, formulae and vocabulary", along with historical context should contribute to dating an inscription. Meiggs and Lewis (1969:81) re-affirmed their commitment to letter-form dating. Mattingly (1976) puzzling over a phrase contended (see below) that the "earliest allowable dating" (1976:44) for the Egesta inscription was Epameinōn (429/8 BCE).

Recently Moroo (2002) has concluded that "there were no rules for using a particular letter form" and they can only "establish general trends". However as Moroo looked at sigmas being used as currency signs for staters, Clinton's (1996) comment probably applies, viz, that stater signs are in "a different domain and therefore irrelevant".

The argument on letter-form evolution seems to have been terminated by Chambers, Gallucci and Spanos (1990). They have shown (see below) that since Antiphōn was the archon in the Egesta inscription which had three-barred sigmas, there was no mass extinction of the three-barred sigma near 445 BCE.

Argument for the date 418/7 BCE

Mattingly (1963:267f) visited the Egestan region and was impressed by the realisation that the Elymite community would have been quite remote from Athens. In addition, in the 450s Athens had serious enemies, both Persian and Greek, with which to contend. The small, remote Egestan region of Sicily would then have seemed of little importance to Athens in the 450s. However, interest developed in Sicily in the 420s when Rhegion and Leontini formed alliances with Athens. Mattingly argued that 418/7 BCE fitted these circumstances. The date was a few years after the Athenians' first excursion to Sicily in 427/6 BCE "to prevent corn being brought in to the Peloponnese" (Thucydides 3:86-88), and just before the expedition in 416/5 BCE for which Egesta campaigned (Thucydides 6:6).

Mattingly (1976:42) drew attention to the phrase ἐς τὸν νομιζόμενον χρόνον i.e. "at the customary time" (IG i2 19:15). When it appeared in IG i2 85 and was dated before 420 BCE, Mattingly argued that a phrase of this kind would not have been used over an extended period and this put the Egesta alliance in the same time period. This would give Antiphon as the Archon of the inscription.

Chambers, Gallucci and Spanos (1990) gave the history of support for reading Antiphōn on the inscription. This included Smart (1972:130) and Wick (1975:Plate XXIIIb). Wick (1981:118) later asserted that in the inscription a letter rho before " ων " was epigraphically improbable and that Meritt's arguments (despite their intention for the opposite) "guarantee the letter phi rather than the letter rho". He included part of a private correspondence from Balcer who testified that he read ΙΦΩΝ clearly (Wick, 1981:121).

Chambers, Gallucci and Spanos (1990) measured the spaces between letter combinations and also published photographs with laser induced contrast at microcrack edges. Their conclusions were that "[t]he group ]βρον … would be incompatible with the space now occupied by ]ιϕον " (1990:42) and a rounded shape appeared which "could not be a rho and can be interpreted as a phi" (1990:43). Both these methods came out decisively for Antiphōn (see appendices 2 - 4 for comments).

Henry suggested that the vertical of the iota was a "mere scratch" (1992:139) and mentioned statistical fluctuations ( ιχσε , βεια. He agreed that some combinations of letters occupied more space than others (Henry, 1992:141 and 1995:239) but then changed Chambers' "incompatible" to "impossible" (Henry, 1992:142). Now the bare possibility of something happening does not mean that it is probable. Chambers' words, "most probably" (1990: 43) as appealed to by Henry (1992:142), rather than admitting "some element of doubt" (Henry, 1992:142) really expressed "little doubt", as indicated by Chambers further down the page.

Henry further argued about the qoppa shape of the "phi", , (Henry, 1995:238) but uneven wear on the marble could have accounted for the apparently unusual length of the vertical line in the phi. Henry made a good point about Chambers' poor "statistics", (Henry, 1995:239) but he does not present any real statistical arguments in reply.

Trêheux (1991:469) seems to have had the last word when he concluded that one should read and restore Antiphōn to the Egestan inscription giving the date as 418/7 BCE. Note that Henry (1998:45) translated this, with emphasis, as "must" but did not support it.


Thucydides' silence on the alliance

Thucydides (6.6.2) has recorded that in the year 416/5 BCE: "the Egestans reminded the Athenians of the alliance made in the time of Laches [427 BCE], during the war in which Leontini was concerned" [15].

Meiggs (1972:599), Bradeen and McGregor (1973:78), and Henry (1992:145,146) all question why Thucydides would have the desperate Egestans appealing to an alliance made with the Athenians ten years before, but remain silent about a nearer, hence more important, alleged alliance made with them in 418/7 BCE.

Chambers, Gallucci and Spanos (1990) survey the different interpretations of Thucydides difficult construction in their appendix and argue reasonably, that the alliance or "miltary collaboration" (1990:52) spoken about could have been between Athens and Leontini. That is, the Egestans reminded the Athenians of what they had done for the Leontines ten years ago and requested the same kind of help. This explains the nature of the actual appeal, but it has not answered the question as to why there was no direct appeal to an alliance made two years before.

In surveying Meiggs (1975), he has these comments on Thucydides' silences.
(1) He has cause to "override the silence of Thucydides" (Meiggs, 1975:96) in accepting evidence from Pausanias on the Stoa Poikile.
(2) He commented that Thucydides gives no details about Oenophyta (Meiggs, 1975:99).
(3) On Oenoe, he wrote "we should not expect such an apparently important event to have left no trace in our literary sources" (Meiggs, 1975:469).
(4) Thucydides is "similarly silent about the capture of Troezen" (Meiggs, 1975:470).

Thus, Meiggs is aware that Thucydides' reporting is sometimes problematic. This was not new because Finlay had already commented that Thucydides, when selecting events to report, applied "his personal canons of relevance" (Finlay, 1972: Introduction 23), mixing extensive minutiae with "astonishing gaps and silences" (Finlay, 1972: Introduction 24). Accordingly, Thucydides silence about a 418 BCE alliance with Egesta is problematic but not without precedent.

Chambers' (1993:174) suggested that Thucydides' exile could have restricted his information. That is Thucydides either, did not know about the alliance of 418/7 BCE or, that he was unsure of it. Chambers noted also that the same problem exists for an alliance date in 458/7 BCE.

Egesta and Athens: consequences

Ridgeway (1888:180) noted that Thucydides presented the Egestans as non-Hellenic (Thucydides, 6.12, 7.57). The Egestans however accommodated an Athenian imperialist ambition of conquering Sicily (Thucydides 6.6) by promising needed finances (with the possibility of more), and giving a rationale for their presence in a non-Hellenic environment. The alliance gradually drew the Athenians into an expedition that produced "the most calamitous of defeats" (Thucydides 7.87).

Conclusion

From the work of Chambers, Gallucci and Spanos (1990 and 1993), the archon Antiphōn should be read or restored to IG I2 19,20 (I3 11).
The alliance between Athens and Egesta should be 418/17 BCE.
Letter-form analysis has been shown to indicate general trends without specific cut-off dates.
Thucydides' silence about the alliance in 416/5 BCE is problematic as are some of his other silences. The alliance was a step that drew the Athenians into the expedition against Sicily.



Bibliography


Primary Sources


Diodorus of Sicily, Library, Translation by C. H. Oldfather, Cambridge, Mass.: Harvard University Press; London: William Heinemann, Ltd. 1989.

Thucydides, History of the Peloponnesian War, Translation by Rex Warner, Penguin, London, 1972


Secondary sources


Bradeen, D.W. and McGregor, M.F., Studies in Fifth-Century Attic Epigraphy, University of Oklahoma Press, Norman.

Chambers, M., The Archon's name in the Athens-Egesta Alliance (IG I3 11) , Zeitschrift für Papyrologie und Epigraphic, 98, 1993.

Chambers, M.H., Photographic Enhancement and a Greek Inscription , Classical Journal, 88, 1992.

Chambers, M.H., Gallucci, R., Spanos, P., Athens' Alliance with Egesta in the Year of Antiphon , Zeitschrift für Papyrologie und Epigraphic, 83, 1990.

Clinton, K., Carol L. Lawton, Attic Document Reliefs: Art and Politics in Ancient Athens  - reviewed by Kevin Clinton, Bryn Mawr Classical Review, online at ccat.sas.upenn.edu/bmcr/1996/96.09.21.html

de Gaertringen, F.H. De Foedere cvm Segestanis a.453 EM6568 , Inscriptiones Graecae, Inscriptiones Atticae evclidis anno anteriores, Ares Publishers Inc., Chicago, 1974

Finlay, M.I., Introduction to Thucydides, History of the Peloponnesian War, Translation by Rex Warner, Penguin, London, 1972

Finlay, M.I., Empire in the Greco-Roman World , Greece and Rome, Clarendon, Oxford, 25, 1978

Henry, A.S., Archon-Dating in Fifth Century Attic Decrees: The 421 Rule , Chiron, 9, 1979

Henry, A.S., Through a Laser Beam Darkly: Space-age Technology and the Egesta Decree (IG I3 11) , Zeitschrift für Papyrologie und Epigraphic, 91, 1992

Henry, A.S., Pour encourager les autres: Athens and Egesta encore , The Classical Quarterly, n.s.45, 1995

Henry, A.S., The Sigma Enigma, Zeitschrift für Papyrologie und Epigraphic, 120, 1998

Mattingly, H.B., The Athenian Coinage Decree, Historia, 10, 1961

Mattingly, H.B., The Growth of Athenian Imperialism , Historia, 12, 1963

Mattingly, H.B., Formal Dating Criteria for Fifth Century Attic Inscriptions , Acta of the Fifth International Congress of Greek and Latin Epigraphy, Cambridge 1967, Oxford, 1971

Mattingly, H.B., Three Attic Decrees , Historia, 25, 1976

Meiggs, R., The Dating of Fifth-Century Attic Inscriptions , Journal of Hellenic Studies, 86, 1966

Meiggs, R., The Athenian Empire, Clarendon Press, Oxford, 1975

Meiggs, R., and Lewis, D., Alliance of Athens with Egesta: (?)458-7 B.C., Clarendon Pres, Oxford

Meritt, B.D. and Wade-Gery, H.T, , The Dating of Documents to the Mid-Fifth Century - I, Journal of Hellenic Studies, 82, 1962

Moroo, A., The Miletus Decree: is dating by three-bar sigma true or illusory?, Newsletter 10, www.csad.ox.ac.uk/CSAD/Newsletter/Newsletter10/Newsletter10e.html

Ridgeway, W., Thucydides VI.2 , The Classical Review, 1888

Smart, J.D., Athens and Egesta , Journal of Hellenic Studies, 92, 1972

Vickers, M.,Fifth Century Chronology and the Coinage Decree , The Journal of Hellenic Studies, 116, 1996

Wick, T.E., The Date of the Athenian-Egestan Alliance , Classical Philology, 76, 1981

Wick, T.E., A Note on the Date of the Athenian-Egestan Alliance , Journal of Hellenic Studies, Vol 95

Woodhead, A.G., Greek Inscriptions , Hesperia, 17, 1948



Appendix 1: The Athenian Archons from 528 to 292 BCE

TABLE VI in Bickerman Chronology pp.137,138 (re-formatted)

BCE

 

BCE

 

BCE

 

BCE

 

BCE

 

 

 

460

Phrasikleides

410

Glaukippos

360

Kallimedes

310

Hieromnemon

 

 

459

Philokles

409

Diokles

359

Eucharistos

309

Demetrios

528

Philoneos

458

Habron

408

Euktemon

358

Kephisodotos

308

Charinos

527

Onetorides

457

Mnesitheides

407

Antigenes

357

Agathokles

307

Anaxikrates

526

Hippias

456

Kallias

406

Kallias

356

Elpines

306

Koroibos

525

Kleisthenes

455

Sosistratos

405

Alexias

355

Kallistratos

305

Euxenippos

524

Miltiades

454

Ariston

404

Pythodoros

354

Diotimos

304

Pherekles

523

Kalliades?

453

Lysikrates

403

Eukleides

353

Thudemos

303

Leostratos

522

Peisistratos?

452

Chairephanes

402

Mikon

352

Aristodemos

302

Nikokles

 

 

451

Antidotos

401

Xenainetos

351

Theellos

301

Klearchos

 

 

450

Euthynos

400

Laches

350

Apolldoros

300

Hegemachos

 

 

449

Pedieus

399

Aristokrates

349

Kallimachos

299

Euktemon

 

 

448

Philiskos

398

Euthykles

348

Theophilos

298

Mnesidemos

497

Archias

447

Timarchides

397

Suniades

347

Theniistokles

297

Antiphates

496

Hipparchos

446

Kallimachos

396

Phormion

346

Archias

296

Nikias

495

Philippos

445

Lysimachides

395

Diophantos

345

Eubulos

295

Nikostratos

494

Pythokritos

444

Praxiteles

394

Eubulides

344

Lykiskos

294

Olympiodoros

493

Themistokles

443

Lysanias

393

Demostratos

343

Pythodotos

293

Olympiodoros II

492

Diognetos

442

Diphilos

392

Philokles

342

Sosigenes

292

Philippos

491

Hybrilides

441

Timokles

391

Nikoteles

341

Nikomachos

 

 

490

Phainippos

440

Morychides

390

Demostratos

340

Theophrastos

 

 

489

Aristeides

439

Glaukinos

389

Antipatros

339

Lysimachides

 

 

488

Anchises

438

Theodoros

388

Pyrgion

338

Chairondas

 

 

487

Telesines

437

Euthymenes

387

Theodotos

337

Phrynichos

 

 

486

?

436

Lysimachos

386

Mystichides

336

Pythodelos

 

 

485

Philokrates

435

Antiochides

385

Dexitheos

335

Euainetos

 

 

484

Leostratos

434

Krates

384

Diotrephes

334

Ktesikles

 

 

483

Nikodemos

433

Apseudes

383

Phanostratos

333

Nikokrates

 

 

482

?

432

Pythodoros

382

Euandros

332

Niketes

 


481

Hypsichides

431

Euthydemos

381

Demophilos

331

Aristophanes

 

480

Kalliades

430

Apollodoros

380

Pytheas

330

Aristophon

 

 

479

Xanthippos

429

Epameinon

379

Nikon

329

Kephisophon

 

 

478

Timosthenes

428

Diotimos

378

Nausinikos

328

Euthykritos

 

 

477

Adeimantos

427

Eukles ]\os

377

Kalleas

327

Hegemon

 

 

476

Phaidon

426

Euthynos

376

Charisandros

326

Chremes

 

 

475

Dromokleides

425

Stratokles

375

Hippodamas

325

Antikles

 

 

474

Akestorides

424

Isarchos

374

Sokratides

324

Hegesias

 

 

473

Menon

423

Ameinias

373

Asteios

323

Kephisodoros

 

 

472

Chares

422

Alkaios

372

Alkisthenes

322

Philokles

 

 

471

Praxiergos

421

Aristion

371

Phrasikleides

321

Archippos

 

 

470

Demotion

420

Astyphilos

370

Dysniketos

320

Neaichmos

 

 

469

Apsephion

419

Archias

369

Lysistratos

319

Apollodoros

 

 

468

Theagenides

418

Antiphon

368

Nausigenes

318

Archippos

 

 

467

Lysistratos

417

Euphemos

367

Polyzelos

317

Demogenes

 

 

466

Lysanias

416

Arimnestos

366

Kephisodoros

316

Demokleides

 

 

465

Lysitheos

415

Charias

365

Chion

315

Praxibulos

 

 

464

Archedemides

414

Teisandros

364

Timokrates

314

Nikodoros

 

 

463

Tlepolemos

413

Kleokritos

363

Charikleides

313

Theophrastos

 

 

462

Konon

412

Kallias

362

Molon

312

Polemon

 

 

461

Euthippos

411

Mnesilochos and Theopompos

361

Nikophemos

311

Simonides

 

 


Appendix 2: Laser enhancement

Chambers, Gallucci and Spanos use of digital imaging and lasers (1990 and 1993) was promising news for other worn inscriptions, but repeated arguments over the one set of data is an obvious call for further experimentation.

Laser-generated photographs.

The inaccurate terminology "laser photograph" (Chambers, 1993:172) seems to claim the authority of Science to obscure the subjectivity of art. Chambers uses words such as "laser-generated photographs" (Chambers, 1993:171) and "laser photograph" as though the laser was responsible for the imaging and not just the contrast. The camera that actually generated the "unenhanced" photographs is not described, nor its settings, nor its orientation. "Two strong standing lamps" for illumination (1990:44) are mentioned but not described.
It should first be noted that 1W laser light directly on a surface could produce annealing at the point where microcracks meet the surface. Also, marble could be ablated from the surface. In order to stop the information content of the marble being degraded, it seems implied (but not stated) that safe limits for intensity were demonstrated on similarly carved marble samples (Chambers, Gallucci and Spanos 1990:45).

Coherence, Monochromaticity and Polarisation

The unique properties of laser light were not used, for example, the coherence property was not used because no holographic image of the surface was produced. This would have produced the equivalent of a "squeeze" but more reproducible and transportable.
There was no explanation why laser light was more useful than that from high intensity lamps, monochromatic or otherwise. Also normal polarised (i.e. incoherent) light has been used to show stress patterns inside transparent samples and this technique may have been more useful than laser light in producing surface effects.

Contrast

Microcracks would have been caused by the original chiselling of the marble. It seems that laser illumination through the marble produced optical contrast at the place where contaminated microcracks intersected with the worn exterior marble surface. The depth profile of this contrast was not reported, leading to questions about contrast of microcracks below the surface. However the experiment of carving a letter phi, abrading it and illuminating it, was a good control feature (Chambers, Gallucci, Spanos, 1990:45). It would also have been good to have carved a number of "phi"s, abraded them to different depths and reproduced the results.
Henry (1992) asked about the possibility of internal flaws. However "an amalgam of defects" (Henry, 1992:145) would tend to produce amorphous shadows without any structure. If any structures did appear they would not be likely to appear as annular circular regions (i.e. donut cross-sections). Hence it would be improbable that an interior defect of exactly the required size, shape and position would be present. This could have been completely ruled out by other techniques for detecting interior defects, e.g. X-rays and ultrasonics, but they were probably not available.
The 1W laser's position behind the marble was mentioned and it may have been set up exactly to illuminate the region under consideration and thus accidentally produce some kind of circular interference pattern through the marble. This would need to be considered as maybe two laser beams were used at once, but the 19 cm thickness of the marble (Chambers, 1992:27,n.11) is greater than the coherence length of the laser light, so the resulting pattern was unlikely to be due to interference.

Collimation

Laser light is normally a fine collimated beam, i.e. it has essentially parallel "rays" that form an point source on interaction with a surface. Collimated light would produce sharp shadows that would enhance the clarity of the inscribed text. However it was not explained how the 1mW laser illuminated the front surface (Chambers 1992:27). The marble would need wide field illumination for the camera to cover the inscription. It was unlikely that the laser would have been rapidly scanned in a raster pattern like a TV image (boustrophedon) so how was a large area illuminated by the laser beam?
It seems the 1W laser beam that was directed through the back surface was dispersed through the marble thus eliminating the collimation of the beam.

Shadows

Why was only one imaging device used and arranged normal to the surface? A better experimental design would involve the use of a number imaging devices (e.g. CCD cameras) arranged to capture the sharp shadows of collimated light. The imaging devices and illumination could then be varied in orientation and the data could then be transformed by appropriate processing to produce a three dimensional image on a computer screen that could rotated and magnified.

Intensity

It would seem that the laser light was used solely because of its intensity. Perhaps other other forms of high intensity and or polarised and or collimated illumination could have been used to the same or better effect.



The limit of resolution for a camera film depends on the size of the grains (and also tha camera aperture).  The film speed was not given so this could not be determined.  Although the information was not given, presumably a fine grained film was used.  The low pass spatial filter would tend to blur the grain edges (as expressly stated) [16] and convert the image into an essentially analogue image, i.e. smooth intensity changes at the edges.  The digitisation process would then reverse the process and create steps in these smooth curves which might not be compatible with the original grains, as illustrated by the diagram below.



Modern (i.e. 2004) digital cameras create images in one stage and bypass this problem.  Modern software can also manipulate images as described whether from space or the backyard.


The fact that the published photos were said to be chosen from a range of photographs and manipulated images implies that these were the most suitable for the letter verification proposed.  As reported, such comparisons based on human observation remain subjective  (Chambers, Gallucci, Spanos, 1990:43).



Appendix 4: Measurement presentation and conclusions


It has been often repeated thatthere are lies, damned lies, and statistics, but statistics need to be considered when presenting this kind of numerical data.  Chambers, Gallucci, Spanos, (1990) show poor knowledge of measurement analysis and data presentation.


For example in Chambers, Gallucci, Spanos, (1990:61) there is no indication of the measuring device (possibly vernier calipers) or the measurement technique e.g. measurements at the top/middle/bottom of the letter combination.   They do however imply that their measurements are from the edges of letters and not the middle. There is an example of raw data, 4.12, 4.07, 4.05, 4.08, 4.11 (in cm).  A mean of 4.0866 cm is given for this and a range that implies an accuracy of (4.09±0.02) cm. These data should have been expressed as (4.09±0.03) cm.[17], so the measurements in the table below probably understate the standard deviation.  Reformatting the table gives:


Table 1

 Line

 Implied

 3 ιφον

 4.09±0.02

 9 ιμελ

 4.28±0.02

 11 ιτον

 4.24±0.02

 12 ιτον

 4.05±0.03

 13 ιδον

 4.21±0.02

 14 ιχσε

 4.53±0.02

 14 ιατε

 4.03±0.02

 15 ιζομ

 4.16±0.02

 16 ιπον

 4.25±0.01

 16 ιδαν

 4.13±0.02

 average

 4.20±0.15


The letter combination ιτον in lines 11 and 12 are revealing.  The difference in the two sample means is 4.24 – 4.05 = 0.19 cm.  The standard deviations (precision of measurement) are 0.02 and 0.03 so the difference between the sample means was six to ten standard deviations (0.19/0.02=9.5 or 0.19/0.03=6.3). The results are then not consistent because the separation of the means is far more than three standard deviations.


This means that engraving ιτον twice has produced a variation in width that exceeds the precision of the measurements.  Accordingly, the stated precision of the two examples can be safely ignored and the width of ιτον albeit based on two sets of measurements, can be provisionally assessed as 4.15±0.13 cm.  This is in agreement with the average width of all four-letter combinations in the first table being 4.20±0.15 cm. 


The four-letter combinations given are tightly grouped (all but ιχσε lie within one standard deviation) and this would seem to be the expectation for this stoichedon inscription [18]  Since ιχσε is 2.2 standard deviations from the mean, some other factor is likely to have intervened to produce this result.  Excluding this reading gives 4.16±0.09 cm.


Table 2

Table 1, p.61

4.20±0.15 cm

Table 2, p.61

4.05±0.10 cm

Table 1 + 2

4.16±0.15 cm

Table 3, p.62

4.51±0.13 cm

Table 4, p.63

4.42±0.13 cm


The standard deviations of Tables 3 and 4 are the same at 0.13 cm.  Tables 1 and 2 when combined have a nominally higher standard deviation of 0.15 cm. However this is stongly effected by the anomolous ιχσε and other things being equal, a standard deviation lower than 0.15 cm would have been expected. 


The similarities of standard deviations implies that the sub-groups were an appropriate choice by Chambers, Gallucci and Spanos.  It also attributes good accuracy to the original engraver when determining the positioning of letters and spaces (<3% error).


The entire set of data given has a four-letter combination width of 4.36±0.20 cm, i.e. a wider spead to cover variations in individual letter widths and combinations.


This can be represented pictorially as:



· The mean for each table sub-group lies within one standard deviation of the mean of the entire set.
· The sub-group variations cluster tighly about the mean with 3% and 4% errors.
· The table sub-group standard deviations do not overlap.
Thus by inspection, the different letter combinations from the two tables form non-overlapping sub-populations.


The conclusion is that Table 3 letter combinations would definitely not be expected to fit into Table 1 spaces.
The conclusion of Chambers, Gallucci, Spanos (1990:42) that "[t]he group ]βρον would be incompatible with the space now occupied by ]ιϕον is statistically correct, despite the presentation flaws.


---

Note: there also seems to be a mistake in the units which give the spacings between letters in Chambers sketch (Chambers, Gallucci, Spanos, 1990:45).  The spaces were recorded as 0.098cm, 0.055 cm and 0.054 cm.


The stoichedon unit was given by Woodhead (1948:58) as 1.47 cm by 1.9 cm.  Using Chambers reported numbers the blank space between letters would be only 7%, 4% and 4%. The picture below shows the letters TANE from line 2 of Wick (1995:plate XXIII).  The average spacing between letters is 28% of the stoichedon unit.


Presumably the spaces in Chambers should have been reported as 0.98cm, 0.55 cm and 0.54 cm.



Footnotes

[1] Lolling, H.G., Archaiologikon Deltion, 1891:105-8, cited by Smart (1972:128), Bradeen and McGregor (1973:75) and Wick (1995:187).

[2]Cited in Fornara ATEPW, 81,n.2.

[3]  Ephialtes took some powers  from the Areopagus in 462/1 BCE and gave some to the Boule of 500.

[4] As noted in Meiggs (1975:100).

[5] Raubitschek, A.E., Transactions and Proceedings of the American Philological Association, 75, 1944, p.10 n.3 cited by Meigs (1966:95), Meiggs and Lewis (1969:81), Bradeen and McGregor (1973:75), Chambers, Gallucci and Spanos (1990:38).

[6] Klaffenbach, G., Supplementum Epigraphicum Graecum 10,7, cited by Mattingly (1963:268,n56) and Bradeen and McGregor (1973:75)

[7] Pritchett, W.K., American Journal of Archaeology, 59, 1955, pp.58-9, cited by Meiggs and Lewis (1969:81).

[8] Meritt, B.D., Bulletin de correspondance hell!"nique, 88, 1964, pp.413, 415 n.2, cited by Meiggs and Lewis (1969:81).

[9] As did Supplementum Epigraphicum Graecum 25, no.3, cited by Ehrenberg, 1975:456,n.138

[10] Beloch, K.J., Griechische Geschichte (1916) cited by Ehrenberg  (1975:456,n.138).

[11] See for example Henry (1979)

[12] Kavvadias, P., Archaiologike Ephemeris, 1897, p.179, as cited in Meiggs (1966:87)

[13] Mattingly, H.B.,The Athenian Coinage Decree , Historia, 10, 1961, pp.148-188; The Growth of Athenian Imperialism , Historia, 12, 1963, pp.257-273;  Periclean Imperialism, Ancient Societies and Institutions: Studies Presented to Victor Ehrenberg, Oxford, 1966, pp.193-223; Epigraphically the Twenties are Too Late, The Annual of the British School at Athens, 65, 1970, pp.129-149; Formal Dating Criteria for Fifth Century attic Inscriptions, Acta of the Fifth International Congress of Greek and Latin Epigraphy, Cambridge 1967, Oxford, 1971, pp.27-33; The Language of Athenian Imperialism, Epigraphica, 36, 1974, pp.31-56; Three Attic Decrees, Historia, 25, 1976, pp.38-44

[14] Note that the AHPG Electronic Reserve has incorrectly labelled Meiggs, 86, 1966 as 85, 1965

[15]Translation by Rex Warner

[16] Chambers, Gallucci, Spanos (1990:42) the pictures were defocused, a procedure and that deemphasises granularity 

[17] More properly (40.9±0.3) mm, however centimeters are used in conformity with the original data presentation.

[18] The stoichedon unit is given in Woodhead, (1948:58) as 1.47 cm by 1.9 cm


Peter's Index Peter's Index  Ancient History Home Ancient History Home  top of page top of page  Samos and Athens Samos and Athens 

email Write me a note if you found this useful