Monthly Archives: April 2013

LABORATORY TESTING OF ANCIENT INSCRIPTIONS: METHODOLOGICAL REFLECTIONS

9 April 2013

LABORATORY TESTING OF ANCIENT INSCRIPTIONS:
METHODOLOGICAL REFLECTIONS

Christopher A. Rollston
Visiting Professor of Northwest Semitic Languages and Literatures
George Washington University
Washington, D.C.

INRODUCTION

One of the most auspicious developments in “forgery detection” during recent years has been the development and usage of laboratory methods to attempt to determine deauthenticity. Most welcome, therefore, is the recent report (e.g., the article by Stephanie Pappas on the Huffington Post) which discusses the laboratory tests on the Gospel of Judas. It seems to me that this is very useful and I think that these results bode well for that particular document (especially because National Geographic’s “Standards and Practices” people are among the very best, and they clearly maintained strict protocols with regard to the laboratory analyses of the Gospel of Judas). Moreover, it has also been reported that the “Jesus Wife Papyrus” has been sent off for laboratory tests. I applaud this as well (I should hasten to add, as discussed below, that it would be particularly useful if the actual lab reports, rather than just the results, would be published as well. For example, it would be very useful to see the graphs generated by the spectroscopy).

I would also wish to emphasize at this time, though, that laboratory reports also require an interpreter. Those of us in the humanities sometimes seem to forget this (at least in practice). Here are a few scenarios, most of which are now attested in the laboratory analysis of actual epigraphic objects from the antiquities market (see section II below). Scenario 1: an epigraphic object is sent for analysis and the laboratory finds modern contaminants under the patina. At that point, the hard scientist has to make a decision. She or he could suggest that the piece is a modern forgery. This is entirely rational and I think quite convincing in most cases. However, it would be possible for him or her to propose that the modern contaminants under the patina could be the result of storage and handling practices (and thus they might suggest that the piece is actually ancient). In other words, the hard scientist must make a decision about the best way to understand the laboratory results. Scenario 2: Or again, let us say that someone has a potsherd from the antiquities market with an ink inscription on it and they send it off to a laboratory for thermoluminescence testing. The results for the thermoluminescence test would probably suggest that the pottery was indeed fired in antiquity. The hard scientist might conclude that this inscription is indeed ancient, based on the results of the thermoluminescence testing. However, this could be wrong. After all, a savvy forger would surely use an ancient potsherd and then forge an inscription on it, in a script and language that corresponds very nicely with the region and time period from which the potsherd came. That is, just because the potsherd is ancient doesn’t mean that the inscription is ancient. Scenario 3: Similarly, someone might send a piece of papyrus off for Accelerator Mass Spectrometry testing (an advanced form of carbon dating) in order to determine if the writing on it is ancient. The AMS test might very well indicate that the papyrus is ancient. However, it would not be prudent for a hard scientist automatically to conclude that the text was authentic, as ancient papyrus is something that can be purchased on the antiquities market and a modern forger worth his or her salt would be wise enough to use that as the medium. Scenario 4: Someone might propose that analyzing the ink should be considered decisive. This is difficult, as finding enough carbon in ink is very difficult (i.e., even for an AMS test, which requires less Carbon 14). But someone might suggest that spectroscopic analysis of the ink could be used, so as to determine the chemical composition of the ink. I believe that this could be useful, at least at times. But I would also hasten to add that enough is known now about the production of ink in antiquity (and the chemical makeup of many inks) that a savvy forger could also find some ancient carbonized remains (e.g., a piece of carbonized timber found on an excavation) and use that as the basis for the production of ink that appears ancient indeed, employing also data known about the rest of the chemicals often found in actual ancient inks (e.g., the right amount of iron, etc.).

For these sorts of reasons, I believe that caution should be the modus operandi. Of course, it is often suggested that forgers are not really that gifted, that savvy, and it is also suggested at times that no one who actually had such knowledge and talent would ever produce modern forgeries (i.e., someone that smart would not do it because of ethical reservations about such activities). Actually, I am convinced that the best modern forgers have formidable intellects and abilities and I think that the recent past demonstrates this rather convincingly, as some modern forgers have fooled some very gifted scholars. Moreover, I find it to be sweet that someone would assume that no one capable of producing really good modern forgeries would do so, but I am not convinced that brains and ethics necessarily always go hand in hand. Thus, I continue to be convinced that the future will see epigraphic forgeries that are absolutely undetectable. Indeed, I have already seen some inscriptions (and analyzed them microscopically) that are virtually perfect, from the script and language to the medium and the patina. I suspect that this will only increase in the coming years.

In addition, I must also admit that I think the field of epigraphy should be very careful about determining the credentials for the laboratory and the hard scientists. Not all labs are equal and not all hard scientists are equal. And there is always an interpretive component. Furthermore, I think that it should go without saying that those with a vested interest in the results should have no part in the sending of pieces to laboratories.

The remainder of this blog post focuses on protocols for laboratory testing as well as a discussion of some laboratory tests which have been performed on epigraphic objects during recent years, including some of the problems with some of these tests. Most of the discussion that follows first appeared in print in an article of mine published a decade ago in the journal Maarav (2003) [footnote 1]. In any case, the focus will be on (1) useful protocols for laboratory testing, (2) followed by concrete discussion of lab testing, based on published lab reports (and some of the problems with these). The hope of all this is that the field can continue to become more and more conscious of the need for the employment of strict protocols for laboratory testing and the importance of understanding the interpretive component that is part of laboratory analysis.

I. PROPOSED PROTOCOLS FOR LABORATORY TESTING

Owners, agents of owners, buyers, agents of buyers, dealers of antiquity, and epigraphists are often interested in having laboratory tests performed on non-provenanced epigraphic objects. In addition, these groups are often required to assess the results and significance of laboratory tests. For methodological reasons, I would argue that certain protocols should be adhered to so that the laboratory tests can actually function as valuable tools.

(1) First, laboratories must be demonstrably legitimate institutions, with the capability of performing the appropriate laboratory tests, and with credentialed individuals performing the tests. Because most epigraphists are not au courante with regard to the best scientific laboratories, there is the potential for substandard laboratories to “authenticate” objects without sufficient scientific bases, and for epigraphists to “accept” such results as authoritative. Furthermore, laboratories that have “authenticated” forged objects (e.g., because of incompetence or collusion) are not to be considered the most reliable and authoritative for future tests [footnote 2].

(2) Second, it is prudent for multiple laboratories, performing the same basic tests, on the same objects, to be used; moreover, it is also imperative that said labs do not communicate together regarding the objects. It is preferable for at least two labs to be used, but cases of conflicting results will necessitate the use of additional labs [footnote 3].

(3) It is of fundamental importance that the laboratory tests be double blind. That is, those requesting the test should not know the names of the laboratories or scientists performing the laboratory tests, and the laboratories should not know the names of the owners, potential buyers, dealers, or (ideally) even the name(s) of the epigraphist(s) studying the epigraphs. Laboratory tests that are not double blind are of minimal value, or perhaps of no value at all, because violation of this protocol ultimately compromises the process, either in theory, or in fact [footnote 4].

(4) Generally, no information about the epigraphic object’s potential or probable antiquity (and thus authenticity) should be given. The reason for this is simply that the scientist performing the laboratory tests, if he or she knows the expected or desired results of the test, can factor in various data in particular ways in order to achieve the expected or desired result [footnote 5].

(5) The full report of all tests should be made available in complete form in the formal publication of the inscription. Laboratory tests that violate these protocols, rigorous though the lab tests may appear, are to be considered of modest value, or of no value at all [footnote 6].

(6) With high profile epigraphs or corpora of epigraphs, when it is readily apparent that there are problems with the laboratory tests performed previously, museums, departments of antiquity, and universities will sometimes need to intervene and appoint disinterested specialists to analyze said epigraphs or corpora. Of course, in circumstances such as this it will not be possible for the tests to be double-blind. Nonetheless, it is expected that a full publication of these lab results will be satisfactory [footnote 7].

II.CASE STUDIES OF SELECTED TYPES OF LABORATORY TESTS

Laboratory tests can be of substantial usefulness in “deauthenticating” epigraphic objects [footnote 8]. Because laboratory tests have been widely used in recent years, and because “laboratory authentication” is often considered definitive, the usefulness of laboratory tests, especially the necessary protocols for laboratory tests and the necessary caveats regarding their decisiveness, merits discussion. For the sake of clarity, various specific laboratory analyses will be discussed; however, it should be noted that this discussion is not exhaustive, but rather illustrative. Finally, although I have attempted to make this discussion accessible, it is inevitable that some specialized vocabulary will be used [footnote 9].

A. PATINAS ON MEDIA AND RADIOCARBON DATING OF ORGANIC MATERIALS

The “Moussaieff Ostraca” have been subjected to certain laboratory tests [footnote 10]. The McCrone affirmed that the soil on the ostraca was Middle Eastern and that the chemical composition of the ink was very similar to that of some of the Dead Sea Scrolls. This data is, of course, of limited usefulness, as carbon based ink is not difficult to produce, and the presence of Middle Eastern soil on an object ultimately proves absolutely nothing about the antiquity of an object [footnote 11].

Mikrofokus Oy Lab focused its analysis on a white patina covering the surface of the ostraca. Using a SEM-EDS (i.e., a scanning electron microscope equipped with an EDS), the McCrone lab had determined that there was a micrometeorite or pyrophoric igniter powder (e.g., from a cigarette lighter) contaminating a sample. Moreover, Mikrofokus Oy affirmed the presence of “spherical particles found which seem to be inside the lower layer of the white material” (i.e., under the patina). The lab then noted that the particles were composed of silicon, and that they were identical in form, size, and chemical composition to those commonly found among flyash particles in industrial environments. Mikrofokus Oy suggested that these were indeed modern contaminants, but ultimately proposed that such particles need not mean that the ostraca were forgeries, even though they were found under the patina. Variable handling and storage practices were assumed to be the cause of this anomaly. Moreover, because producing such a patina today would require a special environment, access to calcite crystals of the right size, specialized knowledge, and a long time, (e.g., “years”), the lab concluded that the patina on the ostraca was a natural precipitation product, and that “what is under the white material must then be genuine.” Note that the Mikrofokus Oy lab, however, was able to duplicate the patina on the ostraca quite closely, using sodium silicate, calcium carbonate, and water.

Obviously, Mikrofokus Oy’s interpretation is a possible interpretation of the data generated by their SEM-EDS test, but it is my opinion that this test does not actually generate much confidence. That is, a number of their interpretations of these test results are potentially troubling. For example, the modern contaminants under the patina could justifiably be understood as being there because the patina was produced during the modern period. Indeed, this is perfectly plausible. Also, the Mikrofokus Oy laboratory assumes that the patina is probably not modern because few people have the appropriate chemicals and the ability to produce such a patina. However, within their own report, Mikrofokus Oy stated that they were able to produce a similar patina, within a very short period. The position of Mikrofokus Oy seems to be that if the patina is not easy to reproduce, then it is safe to assume that it was not done in this case on these ostraca. This is not, however, a necessary conclusion, and I would argue that it is not even necessarily a logical conclusion. In short, it is my opinion that although these tests do not necessarily demonstrate that the Moussaieff Ostraca must be disqualified (i.e., considered forgeries), neither is it possible to state in any way that these tests are non-problematic, and authenticate the ostraca.

The Geological Survey of Israel performed SEM-EDS analyses on the patina of the “Ya‘akov Ossuary” [footnote 12] The results showed that the patina is composed mainly of CaC03 (93%) and contains Si – 5.0%; A1 – 0.7%; Fe – 0.3%; P – 0.4%; and Mg – 0.2%. The report notes that there are no modern elements (such as modern pigments), and the patina adheres firmly to the stone. Again, this is valuable information, but it is imperative that one not conclude or assume, on the basis of this evidence, that this entire inscription is ancient. Rather, this test simply demonstrates that this object need not be disqualified on the basis of the chemical composition of the patina. It certainly does not authenticate the patina. Indeed, the report implicitly concedes this point with the words: “no evidence that might detract from the authenticity of the patina and the inscription was found.” This is an important and honest assessment; however, it must be noted that the absence of certain anomalies in the chemical composition of a patina is not the same as a demonstration of the antiquity of a patina [footnote 13].

The Geological Survey of Israel’s official report for the Jehoash Inscription is much more detailed than the one published (initially) for the ossuary, but this official report is replete with some problems, alas [footnote 14]. For example, (1) the report assumes that if the patina on the medium (in this case, stone) essentially “matches” that which is in the grooves of the inscription, the inscription must be ancient. This is problematic, as it is possible to argue that a modern forger had carefully chosen a stone, cleaned it with non-treated water (e.g., water from a pure spring), incised it, and then the stone was buried and a patina “cultivated,” either passively (just letting it develop naturally with the compounds in the soil and rainfall) or actively (i.e., “watering” the soil at times, and adding some “safe” compounds, such as lime, iron, etc.) that might assist in the development of the patina. The result would be a patina on the surface of the rock that reflected a chemical composition that was similar to that of the inscribed area. Of course, Goren has now demonstrated conclusively that the precise patina described by the Geological Survey of Israel could be produced quite readily in a lab [footnote 15]. (2) The Geological Survey of Israel put enormous credence in the fact that within the patina were organic materials that could be dated (on the basis of 14C) to antiquity. Again, this is an interpretation, but because ancient organic materials are readily available on the market (as well as to those associated with excavations), this cannot be considered the interpretation. A clever forger is certainly capable of attempting to dupe a lab technician by augmenting a fabricated patina with some organic material. Moreover, with regard to the Jehoash Inscription in particular, this is now the most convincing interpretation [footnote 16].

During the past fifty years, of course, Carbon 14 tests have indeed been very helpful in assessing the antiquity of organic material at times. For example, Carbon 14 tests (AMS) were performed on various Dead Sea Scrolls, and the results were decisive, demonstrating the antiquity of the scrolls and providing a control on palaeographic analysis [footnote 17]. Moreover, several decades ago, Carbon 14 tests were performed on the “Hebron Documents” and the tests yielded a modern date, a date that has stood, pace Mendenhall [footnote 18]. Nevertheless, several caveats must be kept in mind even with Carbon 14. (1) Variable handling and storage conditions can have an impact on laboratory tests such as Carbon 14. (2) A Carbon 14 test of organic material (e.g., leather, papyrus) does not necessarily demonstrate the antiquity of the inscription, as pieces of ancient leather and papyrus are sometimes found without visible writing, and can be readily used by a modern forger as the “medium” (i.e., material) for a forgery.

B. THERMOLUMINESCENCE TESTING

Thermoluminescence testing is a potentially important means of determining the age of an object that has suffered in some sort of a conflagration. In essence, very high temperatures reset the thermoluminescence “clock.” Hence, fired ceramics (and “burned” bullae) can be subjected to a thermoluminescence test, and an approximate date of firing can be determined. Thermoluminescence testing on non-provenanced objects, however, is generally not nearly as precise as radiocarbon dating, because there are numerous (quite imprecise) variables to calculate. For example, the chemical content of the soil in which an object has been deposited, the proximity of rocks, and even the types of rocks, all impact the thermoluminescence signal. Therefore, non-provenanced objects, and even provenanced objects not accompanied by samples of the soil surrounding the objects, yield only very approximate dates, with a plus or minus of centuries.

Thermoluminescence texts were recently performed on Hebrew bullae that had ostensibly been fired in some sort of conflagration. Namely, during January 1997, a thermoluminescence test on a bulla was performed by Doreen Stoneham at the Research Laboratory for Archaeology and the History of Art in Oxford. The report is important for a number of reasons. For example, it states that the sample was obtained in “powdered form on 8 January 1997” by Doreen Stoneham, and that it was “presumed” to have been taken from a “small fragment of terracotta.” Stoneham concludes that “using standard methods and techniques it is estimated that the material of the sample was “last fired between 1100 and 1800 years ago” and that this is consistent with the “suggested period of manufacture of the object concerned.” The report concludes by noting that the “result is given in good faith; however the laboratory takes no responsibility for financial loss incurred through an erroneous report being given,” that is, with a certain amount of customary legalese [footnote 19]. Obviously, this thermoluminescence test suggests that the “powder” analyzed was ancient [footnote 20].

Ralf Kotalla of the Ralf Kotalla “Laboratory” (the quotation marks are part of the official document’s report) performed a thermoluminescence test on an Iron Age bulla as well. Kotalla’s report states that the bulla was “2700 years old +/- 20% of the overall age,” and even notes that it originated in the “Orient.” In addition, Kotalla notes that the “assumed age on the basis of stylistic characteristics [is] appr. 2700 years” [footnote 21]. These thermoluminescence test results are especially interesting, because Kotalla was told the precise date of this bulla (i.e., based on the script it dates to ca. 700 BCE). For this reason, he was able to factor in numerous variables and produce a date that corresponded perfectly with the desired result. Stoneham’s test yielded an honest result, but Kotalla’s reflects severe bias. Note that I am not necessarily questioning the antiquity of these bullae. Indeed, for “fired” bullae, a thermoluminescence test is potentially helpful [footnote 22].

C. RADIOGRAPHIC ANALYSES

Spectrographic analyses can be very useful as well, for example, in determining the composition of a metal. Two inscribed arrowheads were analyzed at Conservation Laboratory of the Freer Gallery of Art in Washington, D.C. several years ago. Based on the script, the arrowheads appeared to derive from the Late Bronze Age. The precise results of the spectrographic analyses were as follows: Arrowhead 1: CU 78.925%; ZN – ; PB 7.166%; FE 1.878%; SN 11.940%; and AS 0.091%. Arrowhead 2: CU 73.731%; ZN 0.035%; PB 5.055%; FE 0.981%; SN 20.198%; and AS – ; The absence of zinc in the first arrowhead and the minimal amount in the second arrowhead is important, since copper alloyed with zinc (brass) is rarely found in ancient artifacts. However, McCarter has astutely noted that the “antiquity of the arrowhead is no guarantee of the antiquity of the inscription,” and he states further that “there is now reason to believe that forged arrowheads have begun to appear on the antiquities market” [footnote 23]. The Freer also subjected these arrowheads to microscopic examination as well, and this revealed that the corrosion on the surface of the blades is also present in the incised grooves of their inscriptions. It has been argued that this “mitigates strongly against the possibility that either inscription is modern” [footnote 24]. I concur that this information is helpful, and I am not questioning the antiquity of the inscriptions on these arrowheads. Nonetheless, I would note that creating a corrosion on the surface of a metal object (either before or after cleaning the object) is not difficult for the right person, with access to the right chemicals [footnote 25].

Ultimately, because some forgers are quite savvy about laboratory tests (and because lab scientists have sometimes failed to consider the skills of forgers and their associates), such tests are often helpful, but cannot normally be considered decisive in determining authenticity. Obviously, laboratory tests can disqualify an inscription, but this is not the same as authentication.

D. ANCIENT TOOLS

At this juncture, I should also like to note another type of problematic data that is used to argue for the antiquity of an inscription, namely, that of the tools employed to produce it. For example, regarding the Ya‘akov Ossuary, it was noted that there were “no signs of the use of a modern tool or instrument found” [footnote 26]. Again, this is useful information, but it could never be considered decisive. A shrewd forger could certainly have used an ancient tool, ancient worked metal, or modern bronze or iron consisting of the same composition as ancient tools, so as to have produced the entire inscription, or a portion thereof. Methodologically, it is simply not sage to assume that a modern forger would not take such precautions. Indeed, I would argue that a good modern forger would definitely take such precautions, for modest precautions yield large dividends.

In sum, the use of laboratory tests for inscriptions from the market is auspicious, but the labs conducting the tests must be vetted, protocols for the testing must be put in place in every case, and the results of the laboratory tests must be fully published so that they can be scrutinized as well. In short, there is much to be hopeful about, but methodological doubt must be maintained as well.

[footnote 1]. Rollston, “Non-Provenanced Epigraphs I: Pillaged Antiquities, Northwest Semitic Forgeries, and Protocols for Laboratory Tests.” MAARAV 10 (2003): 135-193.

[footnote 2]. P. Kyle McCarter, Jr. has suggested (personal communication) that it could be very useful to submit three epigraphs to the labs: the epigraph of interest, another that is definitely a fake, and a third that is definitely ancient. That is, he is suggesting the use of “control groups,” as it were.

[footnote 3]. Note, for example, that excavators routinely send samples of organic materials to multiple (often two) labs. The point is that using multiple labs can function as a “control” of sorts.

[footnote 4]. The point here is that recently owners, dealers, or agents thereof, have been taking objects to the lab of their choice, verbalizing expected results, and paying for the tests. This is hardly the ideal.

[footnote 5]. With regard to this issue, note the problems with the thermoluminescence tests discussed earlier.

[footnote 6]. Note that the initial report on the Ya‘akov Ossuary published by the Geological Survey of Israel did not contain much of the raw data or summaries that are to be expected of a SEM-EDS analysis. See A. Rosenfeld and S. Ilani, “SEM-EDS Analyses of Patina,” 29.

[footnote 7]. In my opinion, the IAA should be applauded for doing precisely this during 2003 (for the Ya‘akov Ossuary and the Jehoash Inscription).

[footnote 8] The (2003) report from the commission convened by the IAA to do laboratory analyses of the Ya‘akov Ossuary and the Jehoash Inscription demonstrates this point. Needless to say, I am not convinced by those who have criticized the conclusions of the IAA committee. Indeed, some of the attempted “rebuttals of the report,” read like apologias. In any case, for a brief, but cautious and sound discussion of laboratory testing, see also N. S. Fox, In the Service of the King, 30-31.

[footnote 9]. For a cursory review of the methods discussed here, see W. Ashmore and R. J. Sharer, Discovering our Past: A Brief Introduction to Archaeology, 3rd ed (New York: McGraw-Hill, 2000), esp. 159-168. For analysis using a scanning electron microscope, see also my discussion: C.A. Rollston, “Laboratory Analysis of the Moussaeff Ostraca using the Scanning Electron Microscope (SEM) with an Energy Dispersive X-Ray Microanalyzer (EDS),” 8.

[footnote 10]. For a summary of the lab reports, see C.A. Rollston, “Laboratory Analysis of the Moussaïeff Ostraca using the Scanning Electron Microscope (SEM) with an Energy Dispersive X-Ray Microanalyzer (EDS),” 8-9.

[footnote 11]. Regarding the composition of the ink of ancient inscriptions, see A. Lewis, “Report on the Lachish Letters with Remarks upon the Use of Iron Inks in Antiquity,” in The Lachish Letters, ed. H. Torczyner (Tur-Sinai), 138-193. Of course, carbon (of various sorts) is often found on excavations. The production of carbon ink that appears ancient (e.g., on a 14C test) should not be problematic in the modern period.

[footnote 12]. A Rosenfeld and S. Ilani, “SEM-EDS Analyses of Patina Samples,” BARev 28 (Nov/Dec 2002): 29.

[footnote 13]. Note that there are various other problems with this report as well. For example, the report states that the gray patina found on the surface of the ossuary was “found also within some of the letters, although the inscription was cleaned and the patina is therefore absent from several letters.” However, there is no statement of precisely which letters had this patina and which lacked it (an important issue for various reasons, but especially because of the suggestion that the second half of the inscription was from a later hand). Of course, forgers can use the issue of “cleaning” to their advantage, in various ways.

[footnote 14]. S. Ilani, A. Rosenfeld, and M. Dvorachek, “A Stone Tablet with an Ancient Hebrew Inscription Attributed to Yehoash, King of Judea: Archaeometry and Epigraphy,” GSI Current Research 13 (2003): 109-13.

[footnote 15]. Y. Goren, “An Alternative Interpretation of the Stone Tablet.”

[footnote 16]. Goren proposed this in his article “An Alternative Interpretation of the Stone Tablet,” 7. E. Boaretto’s final report for the IAA also states that the carbon material in the patina of the Jehoash Inscription does not indicate authenticity.

[footnote 17]. See G. Doudna, “Dating the Scrolls on the Basis of Radiocarbon Analysis,” in The Dead Sea Scrolls after Fifty Years, ed. P.W. Flint and J. C. VanderKam (Leiden: Brill, 1998), vol. 1: 430-471.

[footnote 18]. G.E. Mendenhall, “Philistine Documents,” 99.

[footnote 19]. This report is printed in the editio princeps of various non-provenanced inscriptions, namely: R. Deutsch, Messages from the Past: Hebrew Bullae from the Time of Isaiah through the Destruction of the First Temple: Shlomo Moussaieff Collection and an Up to Date Corpus (Tel Aviv: Archaeological Center Publications, 1997), 169. Recently, R. Deutsch also edited a Festschrift in honor of S. Moussaieff, a collector of antiquities, and published a number of additional bullae. See Shlomo: Studies in Epigraphy, Iconography, History, and Archaeology in Honor of Shlomo Moussaieff (Tel Aviv: Archaeological Center Publications, 2003).

[footnote 20]. The date yielded by this test are very broad, and considerably later than might have been expected for an Iron Age bulla. Among the reasons for this are (1) the large number of variables that must be calculated, but cannot be done with precision without data regarding the context in which it was ostensibly found; (2) the theoretical possibility that it suffered a conflagration centuries after its date of manufacture; (3) the theoretical possibility that it has been exposed to some sort of irradiation.

[footnote 21]. For the report, see R. Deutsch, Messages from the Past, 168.

[footnote 22]. I should like to propose that epigraphists and laboratories should anticipate that some forgers may attempt to carve bullae (in relief) from ceramics fired in antiquity, and then to patinate the “epigraph.” Of course, it should be noted that for ostraca, thermoluminescence tests are useless (i.e., a test performed on a fired potsherd) as a forger will find ample sherds in the Middle East upon which to pen his or her logia.

[footnote 23]. P. K. McCarter, Jr., “Two Bronze Arrowheads with Archaic Alphabetic Inscriptions,” in Eretz-Israel 26: Frank Moore Cross Volume, ed. B.A. Levine, et al., (Jerusalem: Israel Exploration Society, 1999), 126-127.

[footnote 24]. P. K. McCarter, Jr., “Two Bronze Arrowheads,” 127.

[footnote 25]. That is, a forger could take a “blank” ancient bronze arrowhead, prepare the surface, forge an inscription, and then create a corrosion (or some type of patina). David A. Scott, senior scientist at the Getty Conservation Institute, has stated (personal communication) that in this case, “the patina under the inscription would be different than the patina over the rest of the surface.” That is, Scott argues that it would be detectable. However, he affirms that the arrowhead “would have to be cut” (and then it should be possible “to tell by metallography that the inscription was added and not original to the object).” If the arrowhead was not cut, Scott suggests that “it should still be possible” to make determinations regarding the differences in the patinas (and thus the antiquity of the inscription itself), but he stated that one would then “have trouble describing the process easily.” For more data about these subjects, see “D.A. Scott, Copper and Bronze in Art: Corrosion, Colorants, Conservation (Getty Conservation Institute, 2002). Ultimately, the point is that even a corrosion should not automatically be assumed to be indicative of antiquity. Certain types of corrosions can be created; therefore, sophisticated laboratory tests (sometimes destructive) are needed to determine the depth and nature of the corrosion in order to assess the antiquity of the inscription.

[footnote 26]. A. Rosenfeld and S. Ilani, “SEM-EDS Analyses of Patina Samples,” BARev 28 (Nov/Dec 2002): 29.