Brill, Robert H.

Dr Robert Brill is in the field of archaeological science, best known for his work on the chemical analysis of ancient glass. Born in the United States of America in 1929, Brill attended West Side High School in Newark, New Jersey, before going on to study for his B.S. degree at Upsala College, also New Jersey (Brill 1993a, Brill 2006, Getty Conservation Institute 2009). Having completed his Ph.D in Physical Chemistry at Rutgers University in 1954, Brill was to return to Upsala College to teach chemistry himself until 1960 when he joined the staff of the Corning Museum of Glass as their second research scientist (Corning Museum of Glass, 2009)

The 1960s saw Brill beginning to develop the analytical techniques that would define the early years of his career at Corning, and yet the scope of his interest within glass remained vast. Indeed, 1961 saw Brill pen a letter to Nature with a colleague, that was a ‘bombshell’, according to Newton, in the field of glass-dating (1971, 3). Here Brill suggested that the rather enigmatic weathering crust found to form on buried glass objects over time could be used to date the object in a method rather similar to dendrochronology, using the separate layers of the shiny lamination (Brill 1961, Brill and Hood 1961, Newton 1971). Whilst in dendrochronology the tree-rings account simply for the tree’s annual growth, in the weathering crust on glass Brill suggested the accumulation of a layer of laminate might respond to some kind of annual event of climatic change (Brill 1961). Unfortunately, despite the examples of the method’s successful applications provided by Brill, such as the almost accurate count of 156 layers on a bottle-base from the York River submerged in 1781 and excavated in 1935, the technique largely failed to convince and did not see widespread adoption (Brill 1961, Newton 1971).

Isotope analysis

The most important of these techniques would prove to be Brill’s pioneering application of lead isotope analysis, hitherto used only in geology, to archaeological objects. Brill first presented this idea at the 1965 Seminar in Examination of Works of Art, held at the Museum of Fine Arts Boston, but the first widely published account of the method seems to be Brill and Wampler’s 1967 article in the American Journal of Archaeology. Here, Brill and Wampler outlined how the technique could be used to provenance the lead contents of archaeological objects to lead ore sources around the world, based on the isotopic signature of various leads, which relates them to ‘ores occurring in different geographical areas’ (1967, 63). These different areas have different signatures because they are of varying geological age, something reflected by the individual lead isotopes which form only after the radioactive decay of uranium and thorium (Brill et al. 1965, Brill and Wampler 1967). While the lead isotope ratios used for provenancing are different, they are not unique: areas geologically similar will yield similar lead isotope signatures (Brill 1970). Furthermore, if leads were salvaged and mixed in ancient times, the isotope ratio will be compromised (Brill 1970). Aside from these two limitations, there is little else that could affect the lead isotope reading an object would yield. As such, Brill’s method was greeted enthusiastically and he went on to develop the technique, as well as oxygen isotope analysis, in his 1970 publication. Here he demonstrated how the technique could be used both to classify early glasses and to a certain extent characterize the ingredients from which they were made (1970, 143).

Chemical-analytical Round Robin

In 1965 Brill launched another important innovation in glass analysis, the comparison of interlaboratory experiments in order to verify analytical results (Brill 1965). ‘Originally inspired by a plea from W E S Turner’, according to Freestone, Brill first mooted his idea at the VIIth International Congress on Glass, in Brussels (Brill 1965a, I. Freestone, pers. comm. 2009). It wasn’t until the VIIIth International Congress on Glass in 1968, however, that Brill fully launched his concept of an ‘analytical round robin’, having distributed a number of reference glasses to be tested in different laboratories using a range of current techniques including X-ray fluorescence and neutron activation analysis (1968, 49). When discussing his motive for the experiment, Brill aptly stated: ‘The truth is that the chemical analysis of glasses is a difficult undertaking and still remains in some senses an art’ (1968, 49). By conducting the round robin experiment, Brill hoped the results gathered from different laboratories would help ‘correlate […] earlier results’ and ‘calibrate future analyses in reference to one another’, as well as suggest which out of the analytical procedures used was the most accurate and effective (1968, 49). The results of the round robin were presented at the ‘IXth International Congress on Glass’ in 1971, and showed that, as Brill suspected, there was poor agreement between certain identified elements, and therefore these might be ‘troublesome’ generally across analyses (1971, 97). These included calcium, aluminum, lead, barium & others (Brill 1971). Aside from their correctional potential, the results, from 45 different laboratories in 15 countries, also provided an enormous data set from which, Brill suggested, the participants could ‘evaluate their own methods and procedures against the findings of other analysts’ (1971, 97). At the time, Brill could hardly have suspected that the data would go on to have such great import, but Croegaard’s generation of preferred glass compositions, from statistical analysis of the data, were used successfully by many people until Brill’s own reference guide was published in 1999 (I. Freestone, ‘pers. comm.’, 2009).

The Middle East

Brill made various forays to the Middle East, including accompanying Wertime’s 1968 survey of the ancient technologies of Iran, alongside other great minds such as the noted ceramicist, Frederick Matson (UCL Institute for Archaeo-Metallurgical Studies 2007). In the years 1963-1964, the Corning Museum of Glass and the University of Missouri, following a long history of excavation at the necropolis of Beth She’Arim, conducted an examination of a huge slab of glass, some 2000 years old, that had been languishing in an ancient cistern (Brill and Wosinski 1965). Brill cannot recall who first suggested this slab, measuring 3.4m by 1.94m, could be made of glass, but the only way to test it was to drill a core through its 45 cm thickness and analyse it (Brill 1967, Brill and Wosinski 1965). On analysis of the core, Brill found that the glass was devitrified and stained, and not very homogenous, with a presence of wollastonite crystals throughout (1965, 219.2). Investigation of the manufacture technology required to produce the slab, suggested that in order to produce such a slab of glass, it would have been necessary to heat over eleven tons of batch material, and sustain it at around 1050˚C for between five and ten days (Brill 1967)! His initial interpretation was that the glass must have been heated either from above or from the sides using a kind of tank furnace; a hypothesis that was proven accurate when excavation underneath the slab suggested it had been melted in situ, in a tank whose floor was a bed of limestone blocks with a thin parting layer of clay (Brill and Wosinski 1965, Brill 1967). Brill’s interpretation, that the slab and its surroundings suggest ‘some early form of reverberatory furnace’ was the first suggestion of the use of tank furnaces in early glassmaking (1967, 92). The evidence at Beth She’arim encouraged further innovative thought because whilst the slab represented glass production on a grand scale, no associated evidence for glass working was found. Brill had already suspected that historical glassmaking occurred in two phases, the heavy ‘engineering’ stage when the glass is formed from the batch ingredients and the ‘crafting’ stage when the glass is formed into artefacts (Brill, pers. comm., 2009). These stages could occur in combination at one location, or at two differing locales, and the time span of production after the initial glass melt is highly flexible. For Brill, the idea of this ‘dual nature of all glassmaking’ was ‘crystallized’ at Beth She’Arim, where only the raw glass production was represented, and would be reinforced later by the contrasting evidence, where working was favoured over production, found at Jalame.


One of the on-running projects of the Corning Museum of Glass published the excavation report from their many field seasons at the ancient glass factory in Jalame, in Late Roman Palestine (Brill 1988, Schreurs and Brill 1984). Brill was called upon to conduct scientific investigations of the huge amount of material generated at the site, in order to exploit the full potential of the artefacts; after all, the site was being excavated specifically because of its role as a glass factory (Brill 1988). Of the vast quantity of glass fragments from Jalame, both vessel sherds and cullet, most were aqua and green and all were soda-lime-silica glasses melted in highly reducing conditions (Schreurs and Brill 1984). Where the melting conditions had been increasingly reducing, a ferri-sulfide chromophore complex was shown to have formed, thus changing the bluey-aqua colour of the glass to an olive, or even an amber shade (Schreurs and Brill 1984). Despite these colour variations, Brill’s further chemical analysis showed the vessel glasses to be highly homogeneous in composition, apart from a clear divide where around 40 glasses demonstrated the intentional addition of manganese (Brill 1988). Brill conducted an investigation of the furnace at Jalame, nicknamed the Red Room, in which there was a mysterious absence of glass finds of any kind (Brill 1988). Whilst work at Beth She’Arim had eventually found there to be five firing chambers responsible for heating the one tank, the fragmentary remains at Jalame made it very difficult to interpret the furnace set-up, apart from the fact that they believed there to have been only one firing chamber (Brill 1988).

The Institute of Nautical Archaeology

In the late eighties Brill was to contribute various studies to the Institute of Nautical Archaeology, following the excavation of a number of exciting shipwrecks including the Serçe Liman, and the Ulu Burun (Barnes et al. 1986, Brill 1989). Here Brill’s own technique of lead isotope analysis was to provide a means for provenancing items aboard ship, and thus determine the ship’s origin and her ports-of-call. The excavators of the Serçe Liman wanted to know whether she was Byzantine or Islamic, a complicated question for lead isotope analysis as the lead ores of the Eastern Mediterranean share geographical characteristics and therefore overlap (Barnes et al. 1986). Using 900 lead net sinkers divided into six loose groupings, Brill found groups III, V and VI to be Byzantine, that is with ores found in modern-day Turkey (Barnes et al. 1986). Group I, however, was taken to be most indicative of the ship’s origin; this group contained net sinkers, but also two ceramic glazes and three glass vessels, all sharing virtually identical lead ores with only one isotopic match, ‘an ore from Anguran, northwest of Tehran’ according to Barnes et al. (1986, 7).

Category: 1948 – 1980
Tags: Beth Shearim,Al-Jalama
Industrial Activity: glass making
Material Composition: Glass