学术史回顾 Historiography of Greek BF/RF Pottery

卡尤斯较早提出过一套古代彩绘陶器的分类框架，将其概括为三大类：红绘、黑绘，以及他所谓的"埃及式"（Cook 1997: 278）。"埃及式"这一类名本身就暴露出 18 世纪分类学的内在张力，他用它指称一种被他视为更原始的彩绘陶，并将其与艺术起源于埃及的观念联系起来（Cook 1997: 278）。这种通过视觉比较区分风格、划分年代阶段的思路，早于 温克尔曼 的艺术史叙事；温克尔曼在《古代艺术史》（1764）中首先讨论艺术的"古老风格"，并把希腊艺术分为四个连续阶段，开启了"风格 = 时间序列"的范式。

Caylus also proposed an early classification of ancient painted pottery into three groups: red-figure, black-figure, and what he called "Egyptian" (Cook 1997: 278). The category name "Egyptian" itself reveals an 18th-century tension: he used it for what he saw as the most primitive painted pottery and linked it to the idea that art originated in Egypt. His comparative-stylistic instinct preceded Winckelmann's art history; Winckelmann's History of Ancient Art (1764) developed it into a four-stage chronological scheme — the "style = time series" paradigm.

同时《古物汇编》深深嵌入了更广泛的巴黎古物学文化之中：在这一文化里，对古代器物与遗迹的研究不仅是学术兴趣的体现，更被视为通向城市共同体与历史自我理解的一种路径（Harloe 2019; Gossman 2019）。

The Recueil was deeply embedded in the broader Parisian antiquarian culture, where the study of objects was seen not just as scholarship but as a route to civic and historical self-understanding (Harloe 2019; Gossman 2019).

• Caylus 1756 vol. II vol. I vol. III vol. V vol. VI vol. VII Gallica
  Caylus, A.-C.-P. de Tubières, comte de. Recueil d'antiquités égyptiennes, étrusques, grecques, romaines et gauloises. 7 vols., Paris, 1752–67.
• "En examinant les précieux restes des Anciens, on peut concevoir une idée sûre de leur goût. Les Arts portent le caractère des Nations qui les ont cultivés."
  Caylus, Recueil II (1756), Avertissement, p. v. Foundational statement of the comparative-cultural-style method later systematized by Winckelmann.
• "Du reste, ces Figures noires, réservées sur le fond, sont pleines d'esprit & de feu."
  Caylus, Recueil II, p. 72 (commentary on Pl. XX). His own technical descriptor — figures noires réservées sur le fond — is the prototype of the BF/RF binomial classification (cf. also pp. 74, 93).
• "材料很快就被分为三大类，红绘、黑绘以及所谓 "埃及式"，其顺序则从风格的精致程度推出。The material was very soon divided into three main classes — red-figure, black-figure and what was called Egyptian — and their order was deduced rightly from the sophistication of their style."
  Cook, R. M. Greek Painted Pottery, 3rd ed., Routledge, 1997, p. 279.
• "The Comte de Caylus was one of the first to analyze in a sound way the differences among Greek, Roman, Etruscan and Egyptian works of art, and as such he was an important forerunner of Winckelmann."
  de Grummond ed. 1997 · Encyclopedia of the History of Classical Archaeology, Routledge, p. 261 (Caylus entry).
• "In 1764 … J. J. Winckelmann gave his sanction to the Greek origin of painted vases, though three years later he conceded to the Etruscans those found in Etruria — not that he had seen any."
  Cook 1997, p. 277. Winckelmann's Geschichte der Kunst des Altertums is the moment the field accepted the vases as Greek; Vulci 1828 finds clinched Athenian provenance.
• "陶片绘制本身就是一项重要创新，因为这种分析法展示了将碎片视为一个组合后，可以具有研究价值的潜力。The drawing of sherds was itself a key innovation, because it demonstrated the potential of the fragment-as-assemblage to bear research value."
  Moser, S. Painting Antiquity. Oxford, 2014, p. 90.

• Morelli 1891
  Morelli, G. Italian Painters: Critical Studies of their Works. London, 1891.
• Beazley Lectures (Kurtz ed. 1989)
  Kurtz, D. ed. Greek Vases: Lectures by J. D. Beazley. Oxford, 1989.
• Hurwit 2015
  Hurwit, J. M. Artists and Signatures in Ancient Greece. Cambridge, 2015. Catalogues the signature corpus underlying Beazley's later judgments. Hurwit on the contemporary signature corpus (citing Vollkommer 2014): "109 Greek potters, 34 vase-painters, and 11 who were both" (Hurwit 2015, p. 96 area). Earliest sculptor's signature: Euthykartides, Delos, c. 625–600 BCE (pp. 3–4).
• "Among Sandro Botticelli's characteristic forms I will mention the hand, with bony fingers — not beautiful, but always full of life; the nails, which, as you perceive in the thumb here, are square with black outlines, and the short nose with dilated nostrils."
  Morelli 1891 ("Principles and Method"), p. 37 — direct demonstration of the morphological method on Botticelli's St. Augustine.
• "The typical, or fundamental, form (Grundform) of hand and ear is characteristic in the works of all independent masters, and affords valuable evidence for identifying them, while mannerisms may, at most, serve to distinguish those of painters wanting in individuality."
  Morelli 1891 (Preface), p. xlv. Introduces the key technical term Grundform ("fundamental form"). Morelli's training was in medical anatomy (Munich 1836, under Döllinger).
• "归因到画工是相对的，因为我们做的归因，是相对于其他归因和我们对希腊陶绘整体的解读做出的判断。Attribution to a hand is relative, because the attributions we make are judgements relative to other attributions and to our overall reading of Greek vase-painting."
  Whitley, J. "Beazley as Theorist." Antiquity 71 (1997): 41.
• "The names of the painters are unknown; but by studying the vases themselves we can get to know their several styles. … Let us begin by saying something about the purpose of these vases; about their shape; and about their technique: and then we can consider the style."
  Beazley, "Attic White Lekythoi," in Kurtz ed. 1989, pp. 26–27. Beazley's clearest one-line creed for stylistic attribution; method order: function → shape → technique → style.
• "It would take some time to argue these attributions; but enough works of both artists have been published for the reader to make up his own mind. … I draw attention to the differences in the rendering of toes, collar-bone, hair, tail."
  Beazley, "Potter and Painter in Ancient Athens," in Kurtz ed. 1989, p. 51. The list of body parts to compare — the direct heir of Morelli's "hands, ears, nails."
• "His lifework culminated in Attic Black-Figure Vase-Painters (Oxford 1956) and Attic Red-Figure Vase-Painters (2nd ed. Oxford, 1963). In the latter alone he identifies almost 800 artists."
  de Grummond ed. 1996, p. 135 (Beazley entry by W. M. Calder III). Beazley's 1910 JHS paper trebled the dozen known works of the Kleophrades Painter; 1911 JHS "revealed an unsuspected master, the Berlin Painter, to whom he at once assigned nearly forty pieces" (Cook 1997, p. 306).
• "Adapting methods that had been used to distinguish Italian Renaissance painters, Beazley attributed thousands of mostly unsigned vases on stylistic grounds."
  Clark, A. J. et al. Understanding Greek Vases, Getty 2002, p. 71 (s.v. "Beazley"). Explicit statement of the Morelli → Beazley methodological lineage.
• Beazley Archive Pottery Database (BAPD)
  Classical Art Research Centre, University of Oxford. The descendant of Beazley's card-files; ~ 95,000 vases as of 2024 (Loy 2025).

Hurwit 2015 直接反驳了 20 世纪末以来对 Beazley 体系的"后人文主义"批评，即认为希腊陶绘的历史不应归于个体陶工，而应被理解为社会与历史力量作用于"可互换的、自动化的"工匠的结果。Hurwit 将这种立场称为"贫瘠"，并以 Vollkommer 2014 的签名汇编（共 154 位有签名的雅典陶工/画师）作为证据反击：陶工的确在器物上签名，且签名率"远高于古代任何其他艺术家"。

Hurwit 2015 directly rebuts the late-20th-c. "posthumanist" critique of Beazley's system — the view that the history of Greek vase-painting should not be attributed to individual potters but instead understood as the operation of social and historical forces on "passive, interchangeable automata." Hurwit calls this position "arid," and answers it with Vollkommer 2014's signature compendium (154 named potters / painters) as evidence that the ancient artisans did sign their work, "more often than any other artist of antiquity."

Hurwit 同时也指出，Beazley 体系仍然支撑着远超风格分析的辩论，例如 Beazley 1910 年命名的 Kleophrades Painter，至今仍是 Alkmaionidai 家族 Megakles 提案的承载者（Hurwit 2015, p. 95）。换言之，"名号"已经长成了一座可继续在其上盖楼的认知结构。

Hurwit also notes that Beazley's system continues to scaffold debates that extend far beyond stylistic analysis — e.g., the Kleophrades Painter named by Beazley in 1910 still bears the proposed identification with Megakles of the Alkmaionidai (Hurwit 2015, p. 95). The "conventional name" has, in other words, grown into a cognitive structure on which further scholarship is built.

• Tonks 1908
  Tonks, O. S. "Experiments with the Black Glaze on Greek Vases." American Journal of Archaeology 12 (1908): 417–27.
• "Careful readings showed that pieces of the black glaze, when submitted to the heat of the furnace, retained their color up to 950° centigrade, but that somewhere between that temperature and 1065° … the black passed over into red."
  Tonks 1908, p. 421. Heating-stage measurements that established the upper temperature bound for the reduced black phase.
• "There was, therefore, three times as much ferrous iron in the glaze as in the clay."
  Tonks 1908, p. 423. The ratio that anchored his pre-firing-recipe model: he proposed adding ferrous oxide to the slip recipe.
• "Approximately 1065° and 950° represent respectively the melting-points of gold and silver, and the idea comes to mind that the Greek potter may have used these metals … to regulate the heat of his kiln."
  Tonks 1908, p. 421. The "gold/silver cone" hypothesis — later refuted by Hussong 1928 (p. 41 n. 2) and Noble 1965 (pp. 75–76) on practical grounds (high price of cone metals, geometry-of-melt observations).
• Binns & Fraser 1929
  Binns, C. F. & Fraser, A. D. "The Genesis of the Greek Black Glaze." American Journal of Archaeology 33.1 (1929): 1–9.
• "The essential points of the present study are, first, that the formation of the black color in the glaze took place in the fire and there only, and, second, that while the reducing fire also blackened the body, this was reoxidized and reddened during the cooling."
  Binns & Fraser 1929, p. 5. The canonical statement of the in-kiln-reduction mechanism — overturning Tonks's pre-firing-recipe model.
• "As soon as the kiln attained a visible red color, or at about 600° C., a type of fuel which would produce a dense smoke was used. This was continued until the finishing temperature of about 950° was reached."
  Binns & Fraser 1929, p. 7.
• "Ferrous oxide is not an obtainable substance on a large scale. It is unstable and oxidizes freely on exposure."
  Binns & Fraser 1929, p. 7. Practical-economic refutation of Tonks's "ferrous oxide added to recipe" approach.

Tonks 1908 在 Princeton 与化学家 William Foster、物理学家 George A. Hulett 合作完成的实验复刻系列，是英美文献中第一次以系统性化学实验回答"黑釉如何形成"。他鉴别出 Brongniart 1844 引述的 Salvétat 锰熔剂说不成立，并提出黑色面源于富铁泥 + 钠盐 (硝酸钠) 助熔剂的预混配方。其最终配方："1 份高岭土加 8 份硝酸钠烧结 → 2 份此烧结物混合 1 份氧化亚铁"（p. 424–25）。但 Tonks 仍把"还原"过程置于配方之中，而非窑内气氛之中。

Tonks's 1908 experimental series at Princeton — done with chemist William Foster and physicist George A. Hulett — was the first systematic Anglo-American laboratory attempt to answer "how does the black glaze form?" He refuted Salvétat's manganese-flux hypothesis (Brongniart 1844) and proposed instead that the black derives from a recipe of iron-rich clay + sodium-nitrate flux. His final recipe: "8 parts of nitrate of soda to 1 of clay, fritted together, and then mixed in the proportions of two parts of frit to one of ferrous oxide" (pp. 424–25). But Tonks still located the "reduction" inside the recipe, not inside the kiln atmosphere.

Binns 是 Alfred University 陶瓷学院的奠基院长，Fraser 是弗吉尼亚大学考古学家。他们 1929 年的论文的核心创新就是把还原从化学配方移到窑内气氛，具体说，是把还原气氛作为一段烧成阶段来管理。他们以一片 Met 博物馆的莱基托斯颈部碎片（cat. no. 22-139-76）为参照，在 Alfred 烧成柜中复刻 ORO 全程：600°C 起进入烟雾还原阶段，950°C 收火，缓慢冷却至 850°C 重新进风。冷却后碎片完全无法与古代原件区分。

Binns was the founding director of the Alfred University ceramics school; Fraser was an archaeologist at the University of Virginia. The crux of their 1929 paper was the shift of "reduction" from chemical recipe to kiln atmosphere — specifically, from a pre-firing additive to a managed firing stage. With a Met. Mus. lekythos-neck fragment (cat. no. 22-139-76) as reference, they ran the full ORO cycle in the Alfred kiln: smoke-reduction from 600°C, finish at 950°C, slow cooling to 850°C with reintroduction of air. The cooled fragments were visually indistinguishable from the ancient original.

这条 Tonks → Binns/Fraser 的方法论纠错链，是整部技术研究史最干净的一次"以实验否定假说"——后来 Aloupi-Siotis 用同样的逻辑否定 Noble 的六偏磷酸钠（Calgon）解絮路线。

This Tonks → Binns/Fraser methodological correction is the cleanest "experiment-refutes-hypothesis" episode in the discipline's history. Aloupi-Siotis would later repeat the same logic in refuting Noble's sodium-hexametaphosphate (Calgon) peptization route.

• Hussong 1928
  Hussong, L. Zur Technik der attischen Gefässkeramik. PhD dissertation, Heidelberg, defended 25 February 1928 under Ludwig Curtius. Published Trier: Jacob Lintz, 1928. Experiments conducted at the Karlsruhe Majolika-Manufaktur.
• "Der tiefschwarze Glanzüberzug der antiken Gefäße, der am besten mit Glasur bezeichnet wird, ist keine einmalige Erfindung."
  Hussong 1928, p. 38. Cf. p. 38: "Die gute gebrannte Glasur besitzt etwa den Härtegrad von Feldspat (H6) und ist vollkommen wasserundurchlässig und säurefest." — Mohs hardness 6, fully water-impermeable, acid-resistant.
• "Die normale Endtemperatur im attischen Ofen [dürfte] etwas darunter liegen, also auf etwa 850° C. … Die beiden Hauptursachen der Verfärbung der antiken Schwarzglasur sind einmal ungenügend starker Auftrag der Glasur, zum andern zu große, meist nur lokal einwirkende Hitze beim Brennen."
  Hussong 1928, pp. 42–43. Hussong's refiring ladder of an originally black sherd (Heidelberg University Collection, p. 41): "at ca. 900 °C, still very dark brown-black; at 920 °C, very dark brown; at 940 °C, dark brown; at 1000 °C, brick red; at 1140 °C, violet-red, partly with matt slate-gloss." This is the canonical empirical scale for "what temperature destroys the black."
• "Experimente im Sinne mittelalterlicher Alchimie lagen dem antiken Töpfer fern."
  Hussong 1928, p. 45. Programmatic rejection of "alchemical" recipe hypotheses (incl. Tonks's gold/silver cones at p. 41 n. 2).
• "Schumann 1942 的复制，是技术研究学界公认的分水岭。The turning point in the technical study was the reproduction of the ancient black glaze in 1942 by Dr. Theodor Schumann."
  Noble 1965 (Preface), p. v. Noble's explicit recognition of Schumann 1942 as the discipline's watershed; he notes Schumann's work "did not become available outside continental Europe until after 1945."
• Schumann, T. "Oberflächenverzierung in der antiken Töpferkunst: Terra sigillata und griechische Schwarzrotmalerei." Berichte der Deutschen Keramischen Gesellschaft 23 (1942): 408–26. (Not in local folder; verbatim summary in Noble 1965 pp. v, 31–37 and Aloupi-Siotis 2008 pp. 54–6.)

Hub · Persons gallery Chaîne · Sources tab Course · Session 1 (Historiography)

对诺布尔六偏磷酸钠方案的否定，伴随着一种更积极、可操作的替代解释：黑光陶衣所需的胶体级悬浮液，很可能主要依赖原料黏土的矿物学性质与水洗分级操作，而不是现代化学分散剂。这一转向促成了 Aloupi-Siotis 等关于"自发分散"机制的实验确认，即 CaO ≤ 2% 的伊利石黏土在水中可自发胶溶，无需添加任何化学剂。

The rejection of Noble's hexametaphosphate route came with a positive alternative: the colloidal suspension required by black-gloss slip was likely produced primarily through the mineralogy of the raw clay plus levigation, not modern chemical dispersants. This turn led to Aloupi-Siotis's experimental confirmation of a "spontaneous dispersion" mechanism — illitic clays with CaO ≤ 2 % self-peptize in water without any added chemical.

在这一框架下，古代陶衣制备的问题就是精确控制物理分级流程的问题：陶工或画师通过水洗沉淀（levigation），分离掉较重颗粒，再通过蒸发等方式把悬浮液浓缩到画线所需的粘度。这条线索后被 Aloupi-Siotis (2008, 2020)、Chaviara 等 (2014, 2016, 2018) 在 THETIS 工坊以及希腊各窑炉遗址的复刻实验中反复验证。

• Noble 1965
  Noble, J. V. The Techniques of Painted Attic Pottery. New York: Watson-Guptill, 1965 (rev. ed. 1988).
• Aloupi-Siotis 2008
  Aloupi-Siotis, E. "Recovery and Revival of Attic Vase-Decoration Techniques." BMSAES 13 (2008): 51–66.
• Walton & Mehta 2015
  Walton, M. et al. "Zn in Athenian Black-Gloss Ceramic Slips: A Trace-Element Marker for Fabrication Technology." JAS 56 (2015): 433–8.
• "In one pint of distilled water dissolve 2½ grams of Calgon (sodium hexametaphosphate), (NaPO₃)₆. Calgon is the peptizing or deflocculating agent."
  Noble 1965, p. 37. The Calgon recipe that became orthodox for thirty-some years.
• "The deflocculating agent used by the ancient Attic potters may have been potash (potassium carbonate, K₂CO₃) produced by soaking wood ashes in water. Potash is completely satisfactory in the preparation of the black glaze."
  Noble 1965, p. 37, n. 7. Noble's own footnote already acknowledged the wood-ash potash alternative — but Calgon remained the textbook recipe in the body of the work.
• "Attic pottery, therefore, was fired at temperatures between 800° and 950°C."
  Noble 1965, p. 76. The canonical firing-temperature range that the chemistry of the next half-century would refine to 890–950 °C (Aloupi-Siotis 2008, 2020).
• "The use of sodium hexametaphosphate (Calgon) as a deflocculating agent is not relevant to Attic BG."
  Jones 2021, p. 96. The mature consensus rejecting Noble's Calgon recipe as a modern laboratory convenience that does not match the ancient Attic chemistry.
• "没有任何六偏磷酸钠或类似添加剂的化学指纹在阿提卡古样中被识别。No chemical fingerprint of sodium hexametaphosphate or analogous additives has been identified in ancient Attic samples."
  Aloupi-Siotis 2008, p. 56.

Schumann 1942 论文中最早提议的解絮凝剂选项实际上包括"葡萄酒沉渣、尿液与某些土壤"（Noble 1965, p. 33 n. 1 引）。Noble 选择以六偏磷酸钠（Calgon）作为标准，这是 1965 年北美博物馆与陶艺学校最易获得且最稳定的实验室化学品。但 Hampe 与 Adam Winter 在德国的复刻尝试反而无法以尿液成功解絮阿提卡黏土，已经预示了 Noble 选定的化学路线本身或许并不对应古代工艺（Noble 1965, p. 33, n. 1）。

Schumann's 1942 paper had actually listed three deflocculant options: "dregs of wine, urine and certain earths" (Noble 1965, p. 33, n. 1). Noble's choice of sodium hexametaphosphate (Calgon) was a practical-experimental decision: it was the most accessible, stable laboratory chemical in 1965 North American museums and pottery schools. But the German replication attempts by Hampe and Adam Winter could not deflocculate Attic clay with urine — already a hint, three decades before Aloupi-Siotis, that the chemical route Noble standardised was probably not the ancient one (Noble 1965, p. 33, n. 1).

Schreiber 1999 默默地推翻了 Calgon：她以"泡过木灰的碱性水 + 阿提卡黏土，比例约 20:80"（p. 53）来制备陶衣，等于把 Schumann 的原始建议悄悄复原。同年，Schreiber 也开始把"black glaze"用"gloss"取而代之，她的理由是阿提卡陶衣含 SiO₂ 不足以构成真正玻璃，且烧成温度低于真正玻化所需（Preface pp. xii–xiii）。2016 年 Chaviara & Aloupi-Siotis 又把"gloss"改回"glaze"；2020 年 Aloupi-Siotis 最终提出"Fe-based glass-ceramic coating (BGc)"，作为材料学上更准确的术语。

Schreiber 1999 quietly overturned Calgon: she prepared the slip with "about 20% clay to 80% alkaline water" (p. 53), where the alkalinity came from soaking wood ashes in water — effectively restoring Schumann's original suggestion. In the same volume Schreiber also began to replace "black glaze" with "gloss", on the explicit grounds that the Attic surface "contains insufficient silica to justify the term glaze, and the firing temperature is not high enough to create the fusion that characterizes a true vitreous glaze" (Preface, pp. xii–xiii). Chaviara & Aloupi-Siotis 2016 swung the terminology back to "glaze"; in 2020 Aloupi-Siotis proposed the final, materials-science-accurate term "Fe-based glass-ceramic coating (BGc)".

所以单单"解絮凝剂用什么"这一个问题，就完整呈现了：(i) Schumann (1942) 给出三种候选，(ii) Noble (1965) 把六偏磷酸钠树为正典，(iii) Schreiber (1999) 悄悄换回木灰，(iv) Aloupi-Siotis (2008) 通过 800+ 次实验否决 Calgon，(v) Jones (2021) 在Advances in Archaeomaterials中正式定案——"Calgon 与阿提卡黑光无关"。

A single question — "what was the deflocculant?" — thus rehearses the whole research history in microcosm: (i) Schumann 1942 lists three candidates; (ii) Noble 1965 canonises hexametaphosphate; (iii) Schreiber 1999 quietly restores wood ash; (iv) Aloupi-Siotis 2008 refutes Calgon via 800+ test firings; (v) Jones 2021 officially closes the chapter — "Calgon is not relevant to Attic BG."

• Jones 2021
  Jones, R. E. "The Decoration and Firing of Ancient Greek Pottery: A Review of Recent Investigations." Advances in Archaeomaterials 2.2 (2021): 67–127.
• Solard 2023
  Solard, B., Amicone, S., Aloupi-Siotis, E. et al. "Back to Black: A Mineralogical and Chemical Characterisation of Atticising 4th-c. BCE Black-Gloss Ware." Archaeological and Anthropological Sciences 15: 135, 2023.
• "The emigration of Attic potters is supported by a fourth century BCE decree recovered in Ephesus (ca. 25 km north of Priene), which states that the Ephesian politeia (citizenship) was offered to two Athenian potters, the brothers Kittios and Bakchios, on the condition that they produce 'black potteries' and the 'hydria of the goddess.'"
  Solard et al. 2023, p. 20 (citing Gauthier 1985, pp. 150–51 and Kratzmüller & Trinkl 2005). The only verbatim treatment of the Ephesian decree (IEph 1420) in the local scholarship folder.
• "By the fourth century BCE, potters making Atticising vases had the same technological knowledge as potters from Attica, suggesting a direct transmission of know-how through apprenticeship."
  Solard et al. 2023, p. 20 (Conclusions). Petrography of Iasos samples (Ia-1 = Attic; Ia-2, Ia-3 = local), pXRF, µ-XRD², SEM-EDS — all converge: gloss compositions at Manfria (Gela), Iasos and Priene fall within the Attic range with only artisanal-scale variation.
• "The local potters fired RF and BG slightly higher (up to 1000°C) than the Attic examples … producing a gloss layer similar to Attic in quality and chemical composition. … At Monte Sannace … a fine local terra rossa clay for the body, from which a finer fraction was prepared for the BG."
  Jones 2021, pp. 101–102 (citing Mirti et al. 2004b and Mangone et al. 2008). Establishes the polycentric Apulian picture: classic-Attic recipe at some centres (Monte Sannace, ~950 °C; Arpi, 1000–1100 °C), red-engobe innovation at others (4th-c. BCE) compensating for calcareous body clay.
• "There is a tendency in English-language publications to view Apulian red-figure pottery as simply a continuation of Attic red-figure, and this has often obscured important differences between the two. The unfounded assumption that Apulian red-figure vases were made by Greeks for colonial Greeks has diverted attention from the Italic people of Apulia who, in fact, provided the principal markets."
  Carpenter 2009 · "Prolegomenon to the Study of Apulian Red-Figure Pottery." AJA 113.1, p. 27. Carpenter dismantles the Taranto-centric model and the Greek-colonist-market assumption, paving the way for the polycentric archaeometric programme.
• "Fragments of pots found in production contexts at Metaponto, including pieces excavated from the kilns themselves, demonstrate conclusively that some Late Apulian painters associated with the workshop of the Darius Painter operated there."
  Carpenter 2009, p. 30. Material evidence (kiln remains, D'Andria 1975 excavation) for non-Taranto Apulian RF production.
• "The market in Apulia and Lucania – free from Athenian monopoly – surely had to be attractive for all those potters who usually operated independently and autonomously. … Whether these were real Athenians or 'westerners' returning to the motherland, as suggested by I. Giudice Rizzo and F. Giudice, is not a determining issue for this study."
  Serino in Schierup & Sabetai eds. 2014, p. 250. The Sicilian/early-Apulian workshops (Pisticci, Amykos, Sisyphus, Painter of the Berlin Dancing Girl) as the western counterpart to Kittos & Bakchios on the Anatolian side.
• Valavanis, P. D. "Σχόλιο στο IEph 1420." Athenaia, 1997, esp. p. 24. Discusses the Ephesian decree.
• PHI · IEph 1420 (full Greek text) · Dittenberger, Syll.³ 354

考古计量证据指向 4 世纪 BCE 的两种并存技术传统：一种沿用阿提卡配方（细 terra rossa，低钙、富铁、富铝、富伊利石），如 Monte Sannace（约 950 °C 烧成）；另一种是 4 世纪后期的本地创新，在含钙坯土上预施红色化妆土（ingobbio rosso），以补偿含钙土烧后偏浅的红色（Jones 2021, p. 102，引 Mangone et al. 2008）。Arpi 烧成温度更达 1000–1100 °C，比阿提卡略高（Giannossa et al. 2020: 244）。

Archaeometric evidence points to two co-existing 4th-c. BCE technical traditions: (a) the classic Attic recipe — fine terra rossa, low-Ca, Fe-rich, Al-rich, illite-rich — as at Monte Sannace (firing ~950 °C); and (b) a late-4th-c. local innovation that pre-applied a red engobe (ingobbio rosso) on calcareous body clay to compensate for the lighter red of Ca-rich substrates (Jones 2021, p. 102, citing Mangone et al. 2008). Arpi fires hotter at 1000–1100 °C, slightly above Attic norms (Giannossa et al. 2020: 244).

Solard et al. 2023 把这一图景延展到东地中海：Manfria（Gela，西西里）、Iasos（卡里亚）、Priene（伊奥尼亚）三地的 4 世纪 BCE 仿阿提卡黑光在化学与矿物学上几乎与阿提卡正品不可区分，仅有"作坊级别"的小差异。其结论："仿阿提卡黑光不可能仅凭外观模仿生产，必先存在技术知识的迁移" (p. 20)。Kittos / Bakchios 在 Ephesos 的雅典潈陶工授籍记录则为这一迁徙提供了罕见的铭文学佐证。

Solard et al. 2023 extend the picture to the eastern Mediterranean: 4th-c. BCE Atticising black gloss from Manfria (Gela, Sicily), Iasos (Caria), and Priene (Ionia) is chemically and mineralogically almost indistinguishable from Attic, with only workshop-scale variation. Their conclusion: Atticising black gloss "cannot be produced without prior technological know-how" — implying direct knowledge migration. The Kittos & Bakchios decree from Ephesos is rare epigraphic corroboration of that migration.

值得注意的是，Carpenter 2009 早已指出，不存在"Taranto 是 Apulian RF 唯一来源"的考古证据；他直接称这种观点为"未经证实的假设"。一万件已知 Apulian RF 中仅约 20% 有出土记录，其大部分出土于 Italic（Peucetian / Messapian）墓中——Ruvo di Puglia、Ceglie del Campo、Rutigliano、Canosa 才是主要市场，而非希腊殖民地。最有力证据：Metaponto 出土的窑炉碎片直接证实部分 Late Apulian 画师就在那里作业（D'Andria 1975 发掘）。

Carpenter 2009 had already established that there is no archaeological evidence that Taranto was the sole source of Apulian RF — he called the assumption "unfounded." Of ~10,000 catalogued Apulian RF vases, only ~20 % have a recorded find-spot, and the bulk of those came from Italic (Peucetian / Messapian) tombs: Ruvo di Puglia, Ceglie del Campo, Rutigliano, Canosa were the principal markets, not Greek colonies. The decisive evidence: kiln fragments at Metaponto confirm that Late Apulian painters of the Darius Painter workshop operated there (D'Andria 1975 excavation).

Serino 在 Schierup ed. 2014 中进一步给出 5 世纪末叶的传播机制：雅典 Kerameikos 在 425–400 BCE 间生产骤降，把南意与西西里市场打开给了"独立、自主作业"的陶工——Pisticci 画师、Amykos 画师、Sisyphus 画师、Berlin Dancing Girl 画师等"早期南意 RF 创建者"。这些迁徙的工匠正是 Kittos / Bakchios 在另一侧的对应案例。

Serino in Schierup ed. 2014 supplies the late-5th-c. transmission mechanism: Kerameikos output collapses between 425–400 BCE, opening the South-Italian and Sicilian markets to "potters who usually operated independently and autonomously" — the founders of early South-Italian RF (Pisticci, Amykos, Sisyphus, Berlin-Dancing-Girl Painters). These mobile craftsmen are the western mirror-image of Kittos & Bakchios on the Anatolian side.

• Schreiber 1999
  Schreiber, T. Athenian Vase Construction: A Potter's Analysis. Malibu: Getty, 1999.
• Tite 2008
  Tite, M. S. "Ceramic Production, Provenance and Use — A Review." Archaeometry 50.2 (2008): 216–31.
• Cuomo di Caprio 2017
  Cuomo di Caprio, N. Ceramics in Archaeology II. Rome, 2017. Synoptic methodology and the chaîne-opératoire framework for ceramics.
• Kopytoff, I. "The Cultural Biography of Things: Commoditization as Process." In A. Appadurai (ed.), The Social Life of Things. Cambridge, 1986, pp. 64–94.
• Leroi-Gourhan, A. Le Geste et la parole. 2 vols., Paris: Albin Michel, 1964–65.

Hub · Institutions thread Chaîne · Pathways tab Course · Session 2

• Maniatis, Aloupi & Stalios 1993
  Maniatis, Y., Aloupi, E. & Stalios, A. D. "New Evidence for the Nature of the Attic Black Gloss." Archaeometry 35.1 (1993): 23–34.
• Aloupi-Siotis 2008
• Chaviara, Aloupi & Siotis 2016
• Aloupi & Siotis 2020
  Aloupi-Siotis, E. & Siotis, V. "Ceramic Technology: How to Characterise Black Fe-based Glass-Ceramic Coatings." Archaeological & Anthropological Sciences 12 (2020): 191.
• "It is coloured by polycrystalline nanoparticles of magnetite or mixed magnetite/hercynite spinels, dispersed in the amorphous K-aluminosilicate phase doped by Fe²⁺/Fe³⁺. … We suggest the term Black Glass-ceramic (BGc) or Fe-BGc."
  Aloupi-Siotis 2020, p. 1, 8. The definitive replacement of "black glaze" / "gloss" terminology with a materials-science-accurate name.
• "The number of ORO firings performed during this period (June 2000 – Dec 2019) exceeds 800."
  Aloupi-Siotis 2020, p. 6, n. 8. The accumulated experimental base from THETIS.
• "The best quality Attic BG with bluish tint (mid fifth century BC sherd from Kerameikos) presented the lowest amount of Fe³⁺/Fe(tot) ~ 20% (or the highest Fe²⁺/Fe(tot))."
  Aloupi-Siotis 2020, p. 7. The empirically determined oxidation-state target.
• "Intense reduction with simultaneous temperature decrease to 800–830 °C is essential for the partial reduction of Fe³⁺ to Fe²⁺ to take place through the formation of magnetite crystals."
  Aloupi-Siotis 2020, p. 6. Definitive firing-window specification.
• "The black layer, of total thickness about 20 μm, consists mainly of polycrystalline magnetite (Fe₃O₄) particles of dimensions from 0.2 μm down to extremely fine sizes embodied in an amorphous vitreous matrix. On the outer surface of the black paint we discovered for the first time a thin clear glassy film of approximate thickness 0.1 μm which is rich in Al and Fe and poor in Si."
  Maniatis, Aloupi & Stalios 1993, p. 23. The first direct in-situ TEM observation of the nanostructure. The thin Al/Fe-rich, Si-poor glass film is the mechanism of the "mirror shine" of the best Attic gloss.
• "The calcareous clays used for the body could not be the starting material for the preparation of the black paint as has been suggested by some authors. … The raw clay used for the paint must be a specially selected material, perhaps always of the same origin."
  Maniatis et al. 1993, pp. 27–28. The case that body and slip share composition but not provenance; slip comes from a separately selected, low-Ca, illite-rich source.
• "The characteristic of the high-quality Attic black glaze is its black-blue color, the result of nanoscale crystals in the series magnetite–hercynite dispersed in a transparent potash-aluminosilicate glass matrix that attains its bluish hue from the bivalent iron (Fe[II])."
  Aloupi-Siotis 2008, p. 120 (in Lapatin ed., Getty). The summary statement of THETIS's research programme.
• "A dense and uniform micromorphology of the glaze … is connected with a very low CaO content (less than 1.5 percent) and with the use of the finest (colloidal) fraction of the clay suspension."
  Aloupi-Siotis 2008, pp. 118–119. Tightens the CaO threshold from the often-cited ≤ 2 % to a stricter ≤ 1.5 % for "high-quality" glaze.
• "Six different clay-soil locations were identified as adequate to produce good quality BG layers, which correspond to 5 different areas in Attica: Panakton, Mount-Parnes, Mount-Hymettos, Mesogea and Laurium."
  Chaviara & Aloupi-Siotis 2016, pp. 517–518 (JAS: Reports 7). The five identified Attic clay-source areas.
• "The combined presence of a black-brown shade and of star-like cracks appears to be a feature of the clay-sources in the Panakton plateau and the neighbouring Mount-Parnes areas. … It would be tempting to suggest that Exekias used clay-paints procured from the Panakton plateau and Mount-Parnes area."
  Chaviara & Aloupi-Siotis 2016, p. 517. A diagnostic surface fingerprint (star-like micro-cracks) that correlates a specific clay source with a specific (Archaic) painter.
• "Illite is the only clay mineral whose dispersion in water is not affected by pH … and may thus allow spontaneous dispersion of a clay-soil in water."
  Chaviara 2014, p. 58. The mineralogical reason why illite-rich, low-Ca Attic clays don't need Calgon: they self-peptize.
• N.C.S.R. Demokritos THETIS Authentics

19 世纪起术语为"黑釉" (black glaze)，反映当时把它误读为真釉的化学预设。Schreiber 1999 改为"gloss"，因 SiO₂ 含量不足以构成真釉。Chaviara & Aloupi-Siotis 2016 又改回"glaze"——因为根据矿物学新证，材料虽不达硅酸盐玻璃但仍构成致密玻化层。最后 Aloupi-Siotis 2020 通盘解决：以材料学规范命名为 Fe-based Glass-Ceramic coating (BGc)：指其由 K-Al-Si 玻璃基质中分散的磁铁矿—铁铝尖晶石纳米晶组成，是真正的"玻璃—陶瓷"复合体（glass-ceramic composite）。该术语已被 Jones 2021 综述正式认可。

From the 19th century the term was "black glaze," reflecting the chemical assumption that the surface was a true glaze. Schreiber 1999 proposed "gloss" because SiO₂ content is too low for a true glaze. Chaviara & Aloupi-Siotis 2016 reverted to "glaze" — citing the mineralogical case that, while not silicate glass, the layer is nevertheless a continuous vitrified film. Finally, Aloupi-Siotis 2020 resolved the question: the surface is to be named, by materials-science conventions, Fe-based Glass-Ceramic coating (BGc) — a true glass-ceramic composite, K-Al-Si glass matrix with magnetite–hercynite nanocrystals dispersed in it. The terminology has been ratified by Jones 2021.

这一术语演进不仅是命名学问题，它对应着研究对象本身的三次再定义：(i) 19 世纪以为它是釉；(ii) 20 世纪晚期把它视为亚玻化的多孔表面（gloss）；(iii) 21 世纪材料学把它鉴别为有意工程化的纳米玻璃陶瓷复合涂层（BGc）。每次术语换装都伴随分辨率与对工艺理解深度的跃升。

The terminological history is not just nomenclature — it tracks three redefinitions of the object itself: (i) the 19th-century reading as glaze; (ii) the late-20th-c. reading as a sub-vitreous porous surface (gloss); (iii) the 21st-c. materials-science identification as a deliberately engineered nano-glass-ceramic composite coating (BGc). Each term shift accompanies a leap in resolution and in depth of process understanding.

Chaviara 2014（NARNIA 项目）抽取 35 处实验室黏土样本，结合 μ-PIXE、μ-XRF、SEM-EDS 数据，与 46 件考古样本（雅典卫城博物馆、Kerameikos 地铁工地、Pallene 与 Raphina 高速公路工地）进行 PCA 分析。结果识别出五处可成功复刻 Attic BG 的低钙伊利石黏土区：Panakton 高原、Parnes 山、Hymettos 山、Mesogea 盆地（Mavroleuki/Markopoulo 一带）、以及，意外的——Laurium 矿区。Laurium 的加入由 Walton & Mehta 2015 的 Zn 异常证据补充：阿提卡 BG 中 Zn 含量（271–1959 ppm）远高于胎土（<361 ppm），其来源最可能为 Laurium 银矿区的菱锌矿（calamine）地表沉积。

Chaviara 2014 (NARNIA project) sampled 35 lab clays and ran PCA combining μ-PIXE, μ-XRF and SEM-EDS data against 46 archaeological samples (Acropolis Museum, Kerameikos METRO excavations, Pallene and Raphina Attica Tollway sites). Five low-Ca illitic clay zones in Attica replicate Attic BG: Panakton plateau, Mt. Parnes, Mt. Hymettos, Mesogea basin (Mavroleuki/Markopoulo area), and — perhaps surprisingly — Laurium. The Laurium inclusion is supplemented by Walton & Mehta 2015's Zn-anomaly evidence: Attic BG Zn concentrations (271–1959 ppm) far exceed those of bodies (< 361 ppm), pointing to Laurium's calamine-bearing surface deposits.

基于这五点产地的发现，Chaviara 提出一个产业组织假说："陶衣土的生产与流通构成一个独立的行业，供应散布在阿提卡各地的众多陶器作坊"（Chaviara 2014, p. 64）。这一断言把考古计量数据反推回经济组织，而这正是 Sapirstein、Stissi、Loy 后来定量建模的对象。

From these five sources Chaviara proposes an industrial-organisation hypothesis: "the production and distribution of clay-paint constituted a separate trade that supplied the large number of pottery workshops distributed throughout Attica" (Chaviara 2014, p. 64). That claim back-propagates archaeometric data into economic organisation — the same object Sapirstein, Stissi and Loy would later quantify.

• Cianchetta 2015
  Cianchetta, I. et al. "Investigating the Firing Protocol of Athenian Pottery Production: A Raman and High-Resolution TEM Study." JAAS 30 (2015): 666–76.
• Walton 2013
  Walton, M. et al. "Material Evidence for Multiple Firings of Ancient Athenian Red-Figure Pottery." J. Am. Ceram. Soc. 96.7 (2013): 2031–35.
• "The relief line microstructure could only be produced through a separate firing, at a hotter temperature, than the other two decorative features. … This is the first reported material evidence of decoration on an Athenian vessel having been produced in such a sequence of steps."
  Walton et al. 2013, pp. 2031, 2034 (on a Kleophrades-Painter sherd, J. Paul Getty Museum 95.AE.31.2). STEM + FIB-prepared cross-sections show the relief line is hercynite-rich and "more melted" than contour line / background (mostly magnetite + ulvospinel). Revives the 1899 Hartwig multi-firing hypothesis.
• "The evidence for multiple firings is not conclusive, resting as it does on a very few individual cases."
  Jones 2021, p. 105. The cautious mature consensus on Walton's multi-firing thesis: real but rare.
• Chaviara 2014
• "Thirty-six replicate samples were painted with refined illitic clay and fired using a three-stage (oxidation–reduction–oxidation) firing protocol. … Results indicate that the temperature of each of the oxidative steps … is most important for controlling the color, the mineralogical phases and the density of the glosses produced."
  Cianchetta et al. 2014/2015, p. 2038 (Microsc. Microanal. 20, Suppl. 3). Black at 950 / 850 / 900 °C; red at 700 / 850 / 900 °C. "Only few time/temperature combinations for each of the three stages lead to the formation of a black slip over a red body."
• "Their archaeological specimen with the highest-quality bluish-black gloss had the most reduced iron (Fe³⁺ 20%) … This matched their modern replica gloss from the finest clay paint (< 0.3 μm) when fired at 890 °C in ORO conditions."
  Jones 2021, p. 97, summarizing Lühl et al. 2014.
• Lühl, L. et al. "Confocal-XANES and the Attic Black Glaze: The Three-Stage Firing Process through Modern Reproduction." Analyst 139.20 (2014): 5224–36.
• "Notably higher Zn concentrations were found in the slips (271–1959 ppm) than in their corresponding body ceramics (< 361 ppm). … This indicates that the Zn was present in the raw clay material prior to firing rather than it being introduced as a ground water contaminant during burial. … The most notable source of acid mine runoff near Athens with a significant Zn sulfide mineralogy would have been the ancient mining sites of Laurium."
  Walton & Mehta 2015, pp. 430, 433–34 (J. Am. Ceram. Soc. 98.2). The "vitriol hypothesis": Attic slip-clay may have been pre-treated with acid mine runoff from Laurium to flocculate calcite. Tentative but currently the only quantitative explanation of the persistent Zn anomaly.
• "Almost identical Zn and Ce signatures are found not only for black glosses produced in Athens but also in Corinth, and, from across the Adriatic sea, in Etruria. … The Zn signatures could be evidence for shared fabrication methods at each locality and not indicative of the provenance of the raw material."
  Walton & Mehta 2015, pp. 434–35. The chemical case for the technology (not the clay) traveling across the Mediterranean — independent of and parallel to the epigraphic Kittos & Bakchios evidence.
• "Top-grade BG was surely produced without the intervention of these procedures."
  Jones 2021, p. 105. The current sceptical position on the Laurium/vitriol hypothesis: real Zn anomaly, but the highest-quality glaze probably formed by levigation alone.

• Sapirstein 2014
  Sapirstein, P. "Demographics and Productivity in the Ancient Athenian Pottery Industry." 2014.
• Sapirstein 2020
• Stissi 2020
• Loy 2025
  Loy, M. "Professional and Part-Time Potters and Painters: Modelling the Production of Athenian Black- and Red-Figure Painted Pottery." JHS, advance online, 2025.
• "The group has left on average about 8.2 vases for each year of activity. … Makron, the Late Archaic cup-painter, is a notable exception. His painting of over twenty extant pots per year is more than twice the norm."
  Sapirstein 2014, p. 177 (in Oakley ed., Athenian Potters and Painters III). The empirical foundation of the AR = 8.2 vases / painter / year baseline. Cohort of 36 painters with ≥ 150 attributed vases each, jointly responsible for over 7,500 extant vases.
• "The total population should never have exceeded seventy artisans engaged directly in painting vases. … The Athenian pottery industry is much smaller than once believed — perhaps a half or a third of previous estimates."
  Sapirstein 2014, pp. 184–85. The lower-bound estimate that Stissi 2020 will challenge.
• "Douris alone was responsible for more attributed vases than the bottom 140 hands all together, while the top 53 hands produced as many as the bottom 582. … One could plausibly argue for a range of 50–100 employed figure-painters at the apogee of Attic production."
  Sapirstein 2020, pp. 87 (in Hasaki & Bentz eds.). Refines 2014 to a 50–100 "positions" range; total Kerameikos employment 200–300.
• "The digital Beazley archive now contains more than 86,000 Athenian figured pots. … A yearly output of 1,000,000–2,000,000 seems a fair guesstimate, certainly for the period between 550 and 400 BC. … The Beazleyan corpus only represents between 0.0125% and 0.066% of the original output."
  Stissi 2020, pp. 97–98 (in Hasaki & Bentz eds.). The publication-bias correction that pushes painter counts an order of magnitude above Sapirstein.
• "A much less romantic Kerameikos than we may like to recognize, with a large body of 'flex workers', and rather unstable workshop teams, producing large numbers of vessels."
  Stissi 2020, p. 105. The reframing from "named masters" to "flex workforce."
• "Sapirstein risks huge distortion by excluding from his counts any artist for whom there are fewer than 30 attributions, thereby … critically under-estimating the size of the industry. … The numbers for the potting and painting industries, however approximative, only balance if we commit to the possibility … that potting was a part-time and seasonal profession, but that (some) painters would be required to work on painting alone full-time."
  Loy 2025, p. 4. The Monte-Carlo reconciliation of Sapirstein vs Stissi.
• "70–290 full-time painters, with an average of 145 painters: these figures resemble Sapirstein's estimates."
  Loy 2025, p. 11. The headline median number from 100,000 Monte-Carlo samples per calculation: median annual production 570,000 pots, range 384k–1,013k.
• "A critical role of the κεραμεύς extended beyond mere craftsmanship to orchestrating workflows. … There needs to be a separate group of painters steadily chipping away, working at a steady average rate, to keep pace with the potters."
  Loy 2025, abstract + p. 13. The redefinition of the kerameus as project manager rather than (only) craftsman.
• "It is hard to imagine that in such a huge and productive industry pieces revered today as objets d'art, such as fine items by Exekias or the Berlin Painter, could have stood out as truly remarkable against a sea of (so much) background noise."
  Loy 2025, p. 2. The argumentative kicker: if Stissi's numbers are right, the canonical "masterpieces" are statistically improbable as contemporary objects of esteem.
• Open the Athenian Pottery Industry model (interactive)

Sapirstein 2013/2014/2020：以"每位专业画师年产 8.2 件归因陶器"作为常数，36 位有≥150 件归因者构成研究组，共 7,500 件。Sapirstein 据此推算：(i) 阿提卡黑/红绘产业总规模 = 120–200 工匠（不是 Cook 1959 的 ~500 人）；(ii) 全盛期专业画师 = 50–100 人；(iii) Beazley 体系实际上把同一个画师"过度细分"，应合并"近 X" 与"X 风格"为一手。结论：阿提卡是一个高度专业化、家族经营的小作坊集群。

Sapirstein 2013/2014/2020: takes "8.2 attributed vases per specialist painter per year" as a constant; his study cohort is 36 painters with ≥ 150 attributions, jointly responsible for ~ 7,500 vases. From there: (i) total Kerameikos workforce 120–200 (not Cook 1959's ~500); (ii) 50–100 specialist painters at apogee; (iii) Beazley's system over-divides hands — "Manner of X" and "X-like" should fold into the master. Conclusion: a highly specialized family-workshop cluster.

Stissi 2020：从同一 BAPD（彼时 86,000 件）反推年产量为 100 万–200 万件；按这一数字，专业画师数应为 500–1,000，总工匠数 3,000–4,000。Stissi 的关键创新是引入出版偏差校正：BAPD 仅覆盖原始产量的 0.0125%–0.066%，因此 Sapirstein 的 38,830 件统计基础就已经低估。Stissi 倾向于一种"较少浪漫，较多流动"的工坊图像：少数核心高产画师 + 一大批不稳定的"flex 工"。

Stissi 2020: from the same BAPD (then 86,000) infers annual production 1–2 million; therefore 500–1,000 painters; total ~3,000–4,000 craftsmen. Stissi's key innovation is the publication-bias correction — BAPD covers only 0.0125 %–0.066 % of original output, so Sapirstein's 38,830-vase base already undercounts. Stissi prefers a "less romantic, more fluid" workshop picture: a few highly productive masters plus a large unstable "flex" workforce.

Loy 2025：以蒙特卡洛 (100k 模拟/参数) 整合两条路线。输入：BAPD 86,000 件、存活率 0.25–1.00%、出版率 5–15%。中位输出：年产 57 万件 (Q1 38 万、Q3 101 万；低于 Stissi)；陶工 320 (210–550)；专业画师 250–820 (中位 430)。关键洞见：要在 4–10 月烧成季内完成全部装运，画师必须全年工作，但陶工可以季节性：即家庭与奴隶在非烧成期改做绘画。Loy 据此推出"项目经理 κεραμεύς"概念。

Loy 2025: integrates both routes via Monte Carlo (100k samples per parameter). Inputs: BAPD 86,000; survival 0.25–1.00 %; publication 5–15 %. Median outputs: 570,000 pots/year (Q1 384k, Q3 1,013k — below Stissi); 320 potters (210–550); painter count 250–820 (median 430). Key insight: to fill summer shipping windows, painters must work year-round, while potters can be seasonal — i.e. families and slaves switch from forming to painting in the off-firing months. Hence Loy's managerial κεραμεύς.

从认识论上看，三方分歧的本质不在数据，而在参数的多义性：存活率从 0.10% 到 1.20% 之间任何取值都"合理"，但年产估算因此变动 10 倍。这正是 Loy 主张用 Monte Carlo 而非单一定值的关键论证（Loy 2025 §IV; cf. Athenian Industry · Sensitivity panel）。

Epistemically, the three positions differ less in data than in parameter-ambiguity: any survival rate from 0.10 % to 1.20 % is defensible, but it shifts the annual estimate by an order of magnitude. This is exactly Loy's argument for Monte Carlo rather than single-value estimates (Loy 2025 §IV; cf. the interactive Sensitivity panel).

Hub · Economy thread Industry · Painter Count panel Chaîne · Chemistry tab