Cat. 7 Portrait Head of a Young Woman

Made of creamy white Dokimeion marble from ancient Phrygia in Asia Minor (modern-day Turkey), this slightly under life-size portrait depicts an elegant young woman, perhaps in her late teens or early twenties.1 Although her face is idealized, her youth is emphasized by the sensitive modeling of her smooth, rounded face and high cheekbones.2 The wide-set, almond-shaped eyes with heavy lids have incised irises truncated at the top by the upper eyelids, as well as shallowly drilled, round pupils (fig. 7.1). The subject’s forward gaze contributes to her serene, remote expression. The lightly incised eyebrows stand out from the supraorbital ridges, and the small, bow-shaped mouth and rounded chin were sculpted with exquisite skill. The lower part of the nose is partially broken away.

The subject’s smooth facial features contrast with the painstakingly textured carving of the elaborate hairstyle. Over the forehead, wavy locks are parted at the center and pulled back, leaving the ears uncovered (fig. 7.2). Above these locks is a narrow, twisted band of fabric or hair that extends to the back of the head.3 Three tiers of delicately carved plaits encircle the top of the head and tuck into one another in the front and back (fig. 7.3). This hairstyle, which was popular in the early to mid-second century A.D., has been described by scholars as the “turban coiffure.” Here it is completed with rows of waves arranged in equal segments that cover the crown of the head and the base of the skull. These waves evoke the so-called melon coiffure, a style developed in the early Hellenistic period that involved segments of hair running from the front to the back of the head, as seen in an early fourth-century B.C. grave monument in Los Angeles (fig. 7.4).4 In the Chicago portrait, the segments are carved in a heavier, more cursory fashion than the strands of hair within the braids, suggesting that the back of the head might not have been intended to be clearly visible to the viewer (fig. 7.5). Below the head are the remains of the neck, an indication that the portrait was once part of a statuary body or bust.5

Since its arrival at the Art Institute of Chicago in 1960, scholars have debated the date of Portrait Head of a Young Woman. Exhibiting sculptural elements that were employed initially in the second century A.D. and again in the fourth, the portrait has been assigned to dates in both centuries. Stylistic analyses of the subject’s hairstyle and, to a lesser extent, her facial features have been the focus of the arguments regarding the date of this work. A review of the evidence that has been marshaled in support each of these positions, along with some new observations, has led to the current dating of the sculpture to approximately A.D. 130/40.

A Second-Century Date

Since the 1960s, roughly half of the scholars who have studied this portrait head have argued for a date in the second-century A.D.6 These claims have been based primarily on the hairstyle, which, as previously noted, has been identified as an example of the second-century turban coiffure. Characterized by a series of narrow braids that encircle the upper part of the head, this style is thought to have developed in part out of the elaborate female hairstyles of the Flavian (A.D. 69–96) and Trajanic (A.D. 98–117) periods, which frequently included a coil of braids at the back of the head. Whereas this braided feature often took the form of a small bun in the Flavian period, it expanded into a much larger “nest” in the Trajanic period, as seen in a portrait that might depict the deified Matidia (A.D. 68–119) (fig. 7.6), the mother of the empress Sabina.7 The turban coiffure was worn by private women in the Trajanic period and was especially popular in the Hadrianic period (A.D. 117–38), but it might have been worn into the early part of the Antonine period (A.D. 138–93).8 Although imperial women are known to have worn variations on the style, there are no extant portraits that clearly represent an empress wearing it.9

The hairstyle depicted in Portrait Head of a Young Woman bears a clear resemblance to the turban coiffures represented in a number of second-century female portraits.10 Significant attention was paid to the accurate representation of women’s hairstyles in Roman portraiture, no doubt in part due to the association of artfully styled hair with cultus.11 Numerous variations on this basic coiffure have been identified, perhaps resulting from sitters’ individual preferences.12

A close comparison to the hairstyle of the Chicago portrait is found in an early Hadrianic bust of a woman in the Musei Capitolini in Rome (fig. 7.7).13 The subject wears an arrangement of neatly parted waves over the forehead, surmounted by a thin, twisted band of hair or fabric marked by short, diagonal lines, above which sits a turban comprising several rows of braids. Additionally, the hair on the top and back of the head displays melon-coiffure-like waves similar to those of the Chicago portrait (fig. 7.8). In both sculptures the crown of the head is clearly visible above the turban, although in the Chicago work a larger portion of it is seen due to the slightly lower placement of the braids at the back of the head.14 Despite the similarities of the two hairstyles, the carving of the Chicago portrait’s coiffure is of notably higher quality, as evidenced by the graceful rendering of the individual strands of hair in the braids and locks over the forehead. In contrast, the details of the Capitoline portrait’s coiffure are substantially harder and more linear throughout.

The Chicago portrait’s facial features, especially the eyes and eyebrows, are also thought to lend some support to a second-century date. In Roman portraits, the articulation of the eyes through the drilling of the pupils and the incising of the irises appeared as a stylistic feature in the Hadrianic period, around A.D. 130, after which it became a common component of the sculptural repertoire, thus offering a terminus post quem for this work.15 It has been suggested that the treatment of the eyes in the Chicago portrait, with their nearly circular irises that terminate at the lower edge of the upper eyelid, along with the drilled pupils, is characteristic of the second century.16 Moreover, the heavy upper eyelids that extend past the outside corners of the lower lids are associated particularly with Antonine portraiture.17 Finally, the eyebrows, which are distinguished by prominent supraorbital ridges, reflect a slightly softened version of the more sharply faceted treatment of the eyebrows that appears in numerous Trajanic and Hadrianic examples, a feature that is thought to have died out during the Antonine period.18

Comparable elements are found in a portrait bust of a woman named Bilia from the ancient city of Apollonia in the Roman province of Macedonia (in modern-day Albania) (fig. 7.9).19 Much like the subject of the Chicago portrait, this young woman, who also wears an elaborate turban coiffure, is shown with heavy-lidded, deep-set, almond-shaped eyes; large pupils that appear to be nearly circular in form; and prominent eyebrows, complete with delicately incised hairs. Although the subject gazes to her right rather than to the front, both likenesses are characterized by a calm, somewhat distant gaze.

Although the findspot of the Chicago portrait is unknown, certain aspects of the eyes, in particular the round pupils, might be evidence for an origin in the Greek East.20 This, when considered in light of the marble’s origin in Asia Minor, makes it tempting to speculate that the work might have been created in the eastern part of the Roman Empire.21 On the other hand, is also possible that the marble was exported and carved elsewhere; indeed, Dokimeion marble was known to be exported to Rome for use in statuary.22

Regardless of where the portrait was created, the rendering of the eyes and other facial features appears to support a second-century date, specifically after A.D. 130, when the articulation of the pupils and irises became a common feature, and likely by the early part of the Antonine period, given the association of the heavy upper lids with the portraits of this era. Taking into account the additional evidence of the turban hairstyle, which was most popular during the Hadrianic period but is not widely attested in Antonine portraiture, it seems probable that this portrait was created between A.D. 130 and 140.23

A Fourth-Century Date

Portrait Head of a Young Woman has also been dated to the fourth century A.D., albeit to different imperial reigns within that epoch.24 Scholars who have taken this position have suggested that, like a number of other female portraits assigned to the same period, the Chicago work is a product of a late antique revival, in which Emperor Constantine (r. A.D. 306–37) and his successors looked to the styles of earlier, favored dynasties for inspiration.25 However, many of the stylistically closest examples that have been assigned to the fourth (and in some cases fifth) century are marked by debates analogous to that about the dating of the Chicago portrait.26 Consequently, such portraits in fact offer no secure grounds in support of a late antique date. While comparisons might be made to more firmly dated portraits from the fourth century, this endeavor is complicated by two factors: a considerably smaller number of portraits was produced in late antiquity,27 and extant examples have been infrequently found in securely datable contexts.28 Nevertheless, based on the limited parallels that do remain, there seem to be notable differences between the turban coiffures favored in the second century and the version of the hairstyle that was popular two centuries later.

In support of an early Constantinian date for the Chicago portrait, Cornelius C. Vermeule argued that the hairstyle of this sculpture resembles that worn by Constantia (after A.D. 293–c. 330), the half sister of Emperor Constantine, in her posthumous numismatic portrait of A.D. 330.29 Although the details are difficult to discern due to its small scale, this commemorative coin type clearly depicts Constantia wearing a thick braid that encircles her head but does not fully cover it (see fig. 7.10).30 In this regard, it is similar to the turban coiffure of the Chicago portrait, although the latter comprises three narrow braids rather than one thick one. The comparison to this coin type was the sole support for Vermeule’s identification of the subject of the Chicago portrait as Constantia in 1960. His interpretation was accepted soon after by Michael Milkovich, George M. A. Hanfmann, and James D. Breckenridge;31 subsequently the subject was identified more generally as a woman of the Constantinian court.32

Constantia’s coiffure in her numismatic portrait more closely resembles a hairstyle seen in some securely dated fourth-century portraits in the round that is formed by one or two thick plaits encircling the head, often extending down to the nape of the neck.33 Described as the “Zopfkranzfrisur” (braided crown hairstyle), this coiffure is seen in a seated statue with a fourth-century portrait that is thought to depict Helena (c. A.D. 250–330), the mother of Emperor Constantine, in the Musei Capitolini in Rome (fig. 7.11).34 Here the turban is composed of two wide braids, the individual strands of which are cursorily worked, in contrast to the finely carved individual hairs of the braids in the Chicago portrait.

An even closer comparison to the numismatic portrait of Constantia, which illustrates the clear distinction between late antique examples of the turban coiffure and those of the second century, is found in an early fourth-century portrait of a woman from Fulda, Germany, in the Museum Schloss Fasanerie (fig. 7.12).35 In this sculpture, three thick braids encircle the head, wrapping from the top of the head to the base of the skull in a nearly vertical orientation rather than horizontally around the crown. Tongue-shaped locks formed by flattened ringlets cover the forehead, fully conceal the ears, and extend relatively low on the back of the neck, creating a heavy, almost helmetlike style.36

Based on these comparisons, it is evident that such late antique hairstyles, although incorporating one or more braids encircling the head, are nevertheless distinct from the second-century turban coiffure as represented in numerous portraits dated securely to that period. The delicate and elegant hairstyle of the Chicago portrait clearly has more in common with the second-century than the fourth-century examples. While it is unclear whether the late antique coiffures intentionally evoked this earlier style, the fact that the basic form of a turban of braids was repeatedly used in numerous portraits in both centuries seems to suggest that it might have held some general significance, as will be addressed below.

Scholars arguing for a fourth-century date have also called attention to certain aspects of the portrait’s facial features, namely the size of the eyes and the subject’s fixed, forward gaze. In late antique portraiture, the eyes are often enlarged, at times unnaturally so.37 When combined with an intense gaze, such eyes are thought to convey a sense of greater expressivity.38

The eyes of the Chicago portrait have frequently been described as “large.”39 On closer inspection, however, it is apparent that they are proportional to the face as a whole; rather, it is the pupils that appear somewhat generously sized. It is perhaps due to the size of the pupils that the subject has been described as having a “visionary gaze” or a “heavenward glance” of the type often associated with late antique portraits.40 Still, the eyes are not characterized by the same enlarged proportions seen in some fourth-century examples, such as a portrait thought to depict Fausta (A.D. 289–326), the wife of Constantine, in the Musée du Louvre (fig. 7.13).41 In this sculpture, the subject’s heavy eyebrows, which nearly meet at the bridge of the nose, frame oversize, wide-set eyes. Although the eyes similarly include nearly circular irises and large, shallowly drilled pupils, here the large size of the eyes as a whole clearly distinguishes them from those of Portrait Head of a Young Woman. The “late antique gaze” that some have read into the latter may therefore owe more to a subjective interpretation of the emotional communication of the work than to the sculptor’s treatment of the facial features.

Bente Kiilerich argued that the face of the Chicago portrait might have been reworked in late antiquity, given what she saw as the fourth-century rendering of the eyes, along with the soft treatment of the surface and the precision of the carving.42 If the face was recarved, one might expect the hairstyle (provided that it was preserved) to be disproportionately large.43 However, examination of the portrait suggests that the proportions of the face are appropriate to the head as a whole. Furthermore, recarved eyes often appear very deeply set and unnaturally large, which is not the case here.44 It therefore seems most probable that the entire portrait, including the eyes, the rest of the facial features, and the hairstyle, were carved in the second century.

The Turban Coiffure as a Symbol of Virtue

As noted above, the turban coiffure’s characteristic arrangement of plaits encircling the head appears to have developed in part out of the smaller bun or the more sizable nest of braids at the back of the head that was featured in earlier Flavian and Trajanic hairstyles, which also included diadem-like toupets or crests of hair over the forehead (see fig. 7.6). It has been argued that, rather than being simply an elaboration upon the braided features of these earlier styles, the turban coiffure might have been intentionally created to communicate a distinct message.

Numerous scholars have observed that the turban coiffure resembles the hairstyle depicted in portraits of Roman priestesses, particularly the headdresses of the Vestal Virgins, the high-ranking priestesses of Vesta, the Roman goddess of the hearth.45 Recently, Molly M. Lindner has explored this connection, arguing that private women in Italy of nonelite status developed the turban coiffure to emulate the infula that the Vestals wrapped around their heads (see fig. 7.14).46 Viewed as a sign of sanctitas (holiness, inviolability, and sacredness), the infula was also worn by other Roman priests and priestesses, suppliants seeking pardon or protection, and sacrificial victims.47 According to Lindner, following the increased emphasis on morality (especially that of women) under the emperors Domitian (r. A.D. 81–96) and Trajan, nonelite women sought to imitate the Vestals in order to demonstrate their own adherence to moral standards, devotion to the gods, and embodiment of sanctitas (by the second century, sanctitas had become synonymous with the virtue of castitas—chastity, viewed broadly in the context of purity).48 To a lesser extent, the turban coiffure might also have evoked the distinctive hairstyle the Vestals wore beneath the infula, known as the seni crines (generally translated as “six tressed”), which involved rows of braids wrapped around the head.49 This style was worn by both Vestals and Roman brides as a symbol of their castitas, here specifically referring to their virginity.50 For Roman matrons, the allusion to the seni crines might instead have suggested the associations of castitas with marital fidelity.51

Lindner further proposed that the turban coiffure was soon adopted by Roman women of elite (but not imperial) status, some of whom were priestesses of other cults.52 Priestesses gained considerable prominence in their local communities, and in some cases they were honored with portraits in life and in death, which might have incorporated the turban coiffure in order to suggest the infula of the Vestals.53 The association of this hairstyle with portraits of women of priestly (but non-Vestal) status is evident in an over-life-size, second-century portrait of an elderly priestess in the Museum of Fine Arts, Boston, which shows her wearing a turban of five braids that decrease in width as they proceed from her forehead upward; below, the locks over the forehead are parted in the center and brushed to the sides of the head (fig. 7.15).54 The subject’s engagement in a religious act is implied by the palla drawn over her head as a veil; she is identified as a priestess because she is depicted in the act of burning incense, a rare occurrence in Roman portraiture.55 This portrait, which was the only sculpture found in a vaulted tomb in Pozzuoli in Campania, Italy, was undoubtedly a lavish and costly commemoration that alluded to the subject’s elevated status and presumably also her significant public role.56

Based on the use of the turban coiffure in the Chicago portrait, Evelyn B. Harrison hypothesized that it might represent a priestess.57 This hypothesis cannot be verified, due to the absence of the portrait’s statuary body or bust and any accompanying inscription, as well as a lack of information regarding the context in which it was found. Regardless of the subject’s social standing or potential priestly role, it seems reasonable to suppose that her hairstyle of neatly wrapped braids helped communicate a message about her character. Although the motivations behind the development and popularity of the turban coiffure are still not entirely clear, its associations with the headdresses and hairstyles of esteemed women of high virtue, most notably the Vestal Virgins, suggest that it may have been intended to convey specific ideas about the subject, perhaps emphasizing her morality, modesty, and chastity.

Conclusion

Portrait Head of a Young Woman is characterized by a hairstyle and facial features that have long been thought to support a date in either the second or the fourth century A.D. Based on the evidence examined here, the work most likely dates to the second century A.D., specifically to A.D. 130/40. The turban hairstyle can be confidently dated to the second century on the basis of its resemblance to the coiffures of numerous portraits made in the Hadrianic period, many of which have been dated according to other stylistic features, including the articulation (or lack thereof) of the eyes. The drilled pupils and incised irises of the Chicago portrait indicate a date after around A.D. 130, while other aspects of the eyes, such as the heavy eyelids, are characteristic of Antonine portraiture, suggesting that it was made during the transition from the Hadrianic to the Antonine style. Given the general similarities between the turban coiffure and the headdresses and hairstyles worn by women known for their virtue, particularly the Vestal Virgins, it is possible that the subject’s elegantly braided tresses projected a comparable image of her virtuous nature.
Katharine A. Raff

Technical Report

Technical Summary

This object is a fragment of a portrait bust or statue depicting a young woman. The fragment was carved from a single block of exceedingly fine-grained, white marble, marked by exceptional translucency and brilliance, that has been identified as Dokimeion (fig. 7.16). A variety of toolmarks are visible across the surface. No evidence of polychromy or gilding has been detected. Burial conditions have given the stone a slight ocher tone overall, and further evidence of antiquity can be found in the abundance of root marks distributed over the surface, particularly in the braids on the back of the head. As a whole the head is in stable condition; however, it has suffered a number of noticeable losses. Documentation in the curatorial and conservation files indicates that the object has undergone various surface treatments, generally of an aesthetic nature, in particular to mitigate a once-heavier incrustation of rootlets.

Structure

Mineral/Chemical Composition

The marble is a warm, bright white with a slightly ocher hue that is due to the presence of surface deposits in carved areas. In highly polished areas, the translucency of the marble is readily apparent. Vertical striations and veins are visible throughout.

Primary mineral: calcite (calcium carbonate, CaCO₃)58

Accessory minerals: graphite, C (traces)

Petrographic Description

A sample roughly 2.1 cm high by 1.6 cm wide was removed from an area of existing loss at the base of the neck on the proper right side. The sample was then used to perform minero-petrographic analysis. Part of the sample was finely ground, and the resulting powder was analyzed using X-ray diffraction to determine whether dolomite is present. The remaining portion of the sample was mounted onto a glass slide and ground to a thickness of 30 µm for study under a polarizing microscope.59

Grain size: fine (average MGS less than 2 mm)

Maximum grain size (MGS): 0.24 mm

Fabric: homeoblastic mosaic

Calcite boundaries: embayed

Microscopic examination of the prepared thin-section sample revealed some intercrystalline decohesion60 and confirmed the presence of traces of micritic protolith.61

Thin-section photomicrograph: fig. 7.17

Provenance

Marble type: Dokimeion (marmor docimenum)62

Quarry site: Dokimeion, near modern-day İscehisar, Afyonkarahisar Province, Turkey63

The determination of the marble as Dokimeion was made on the basis of the results of both minero-petrographic analysis and isotopic analysis.64

Isotopic ratio diagram: fig. 7.18

Fabrication

Method

The head is a fragment of a portrait bust or statue. The existing fragment was carved from a single block of stone using the various hand tools and implements that would have been customary for the period, such as chisels, drills, rasps, and scrapers.65

The break edge along the neck has a somewhat flat but nonetheless irregular profile, and the simplest explanation for its topography is that the head was broken away without intent from a larger mass of stone. If the portrait was originally a bust, it was in all probability carved from a single block of stone, and the neck would have been a point of particular vulnerability, prone to damage or fracture. If the head was carved separately to be included in a larger sculpture, it would likely have included a projection made up of the neck and perhaps the collarbone, and it would have been finished with a keyed terminus or tenon meant for insertion into a cavity or socket.66 This extension of stone would also have been highly susceptible to impact and damage. It is also possible that the head was removed purposefully. A deep gouge extends across the surface of the break edge from the front of the neck toward the back on the proper right side, near an area of damage toward the front of the neck. It is possible that this gouge is the result of a percussive blow intended to separate the head from the stone below. While it is impossible to say for certain exactly what happened to the head, it is clear, based on the presence of rootlets on the break edge of the neck, that it was broken prior to burial.

Evidence of Construction/Fabrication

Drills of varying sizes were used to create the nostrils, vestiges of which remain visible in the damaged end of the nose; the pupils of the eyes; and the front edge of the proper left ear, as evidenced by the circular depressions at the ends of the carved line (fig. 7.19).67 At the inside corners of the eyes, a drill approximately 1.75 mm in diameter was used to make tear ducts, increasing the naturalism of the portrait and drawing attention to the eyes.

The sculptor used a flat chisel to carve the irises. The proper left iris is somewhat irregular in shape, which may betray a slip of the sculptor’s hand (fig. 7.20). A very fine flat chisel was employed to delineate the many strands of hair and the braids that are wound around the head. To achieve this fine detail, the sculptor used the edge of the tool, held at an angle to the stone. The sculptor appears to have taken much less care in the execution of the waves of hair below the turban at the nape of the neck. This more perfunctory handling of the carving is particularly noticeable directly behind the ears (see fig. 7.21). This suggests that, although the object was intended to be viewed in the round, this area was not a particular focal point.68

Traces of a rasp are visible on the back of the neck (fig. 7.22), a roughness that contrasts with the high level of finish on the face. This provides further evidence that this area of the sculpture was not expected to be noticed. Encrusting rootlets overlying the rasp marks indicate their presence prior to burial.

The philtrum was carved with a roundel (fig. 7.23).

The marble on the face is highly polished, an effect that would have been realized using a slurry of an abrasive powder such as pumice.69

Artist’s/Fabricator’s Marks

No artist’s or fabricator’s marks were observed.

Additional or Applied Materials

Microscopic examination of the stone’s surfaces revealed no traces of polychromy or gilding.70

Condition Summary

The sculpture has sustained several conspicuous losses. Most notably, the end of the nose is missing, and laminar cleavage, vertically oriented in line with the veins, is visible in association with the fracture. The back halves of both ears are also missing. There is a large, ovoid spall, with a smaller lacuna adjacent to it, on the bottom proper right edge of the neck. A lesser but still significant area of damage can be seen on the front edge of the neck. Residues from the burial environment in the form of red-brown material are present on the surface of this loss. There is a sizable loss to the topmost braid on the back of the head.

Elsewhere minor chips, spalls, and lacunae are in evidence, particularly throughout the hair. There is a gouge with associated bruising and friable material on the middle braid on the proper left side of the head. A small chip is present on the outside edge of the bottom proper left eyelid. A spall is visible nearby on the proper left cheekbone.

Root incrustation—desirable evidence of antiquity—can be seen overall, particularly in the braids on the back of the head and in the hair at the nape of the neck (fig. 7.24).

Extremely fine scratches and abrasions are present across the surface of the face. These may be the result of efforts to remove burial incrustations. When the object is examined under ultraviolet radiation, the extent of this scratching is all the more apparent (fig. 7.25).71

A white, chalky material, likely the remains of a plaster support, is present on the surface of the break edge of the neck. A shiny substance, perhaps from a previous intervention, is visible on the face of the large spall at the base of the neck on the proper right side.

Conservation History

A 1964 condition report in the Conservation Department files, documenting the application of a spray coating of a synthetic resin consolidant, references a previous cleaning, although no date or details of the prior intervention are supplied.72 Additional documentation in the curatorial object file reveals that the object was cleaned again in 1976 using an aqueous solution with added detergent. In 1981 further refinements to these cleaning efforts were performed, this time with some amount of mechanical cleaning of the encrusting rootlets.73 Undated collection photographs in the curatorial object file demonstrate that the head was previously much more heavily encrusted with rootlets and that the base of the neck once bore a plaster support, but it is difficult to construct a time line for the removal of these features based on the existing documentation.
Rachel C. Sabino, with contributions by Lorenzo Lazzarini

6) For the second-century dating of the Chicago portrait’s hairstyle, see Harrison, “Constantinian Portrait,” p. 88; Sydow, Zur Kunstgeschichte des spätantiken Porträts, p. 152; Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, pp. 65–66, cat. 86, item o; Fittschen, “Courtly Portraits,” p. 51 n. 42; Davies, “Portrait of a Woman,” p. 173, cat. 128; Vermeule, “Roman Art,” p. 74, cat. 51; Kiilerich, Late Fourth Century Classicism, pp. 116 n. 374, 117 (here the hairstyle is dated to the second century, while the surface treatment and delicate carving of the face as well as the rendering of the eyes are dated to the Theodosian period, A.D. 379–95); Manchester and Greuel, “Marble Portrait Head of a Young Woman” (listing previously assigned dates from the middle of the second to the late fourth century).

8) On the prevalence of the turban coiffure during the Hadrianic period, see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, pp. 62–63, cats. 83–84; Fittschen, “Courtly Portraits,” pp. 42, 44. Klaus Fittschen questioned the extent to which the turban hairstyle was worn during the Antonine period, noting that it might have been worn into the early part of that period but that this cannot be determined with certainty; see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, p. 63, cat. 84; p. 65, cat. 85; p. 66, cat. 86, n. 6. However, other scholars have suggested that it was worn into the Antonine period; see Breckenridge, Likeness, p. 247, fig. 131; Vermeule, Roman Imperial Art in Greece and Asia Minor, p. 365; Breckenridge, “Head of a Lady,” p. 290, cat. 268; Schade, Frauen in der Spätantike, p. 101.

9) Sabina (A.D. 83/86–136/137), the wife of Emperor Hadrian, wore a variation on the turban coiffure; see Fittschen, “Courtly Portraits,” p. 42. This variation is apparent in her most widespread portrait type, in which the wavy locks parted over her forehead are pulled back into a loosely wound, nest-like coil placed on the upper part of the back of her head. For an example, see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, pp. 10–12, cat. 10. It has been suggested that a portrait of a woman in the guise of Venus Genetrix that incorporates the turban coiffure is a portrait of Sabina; see Calza, Ritratti greci e romani, pp. 77–78, cat. 124, pl. 72. However, it has also been argued that this portrait depicts a private woman; see D’Ambra, “Nudity and Adornment,” pp. 107–08. See also Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, p. 62, cat. 83.

21) For a close comparison to the Chicago portrait that is also made of Dokimeion marble, see the slightly under life-size portrait bust of a woman wearing a turban coiffure, which was found in modern-day Chania (ancient Kydonia) (c. A.D. 410; Archaeological Museum of Chania, Crete, Λ 3176), in Tzanakaki, “Bust of a Lady.” Much like the Chicago work, this sculpture has also been dated variously from the second century through late antiquity because of its hairstyle and the sensitive rendering of its facial features. On the basis of the use of Dokimeion marble and the high quality of the carving, it has been suggested that the portrait might have been imported into Crete following its creation in a workshop in Asia Minor, specifically in Aphrodisias. Could the Chicago portrait have been produced under similar circumstances? For an overview of the quarries of Dokimeion and their importance to the Roman marble trade, see Long, “Urbanism, Art, and Economy,” pp. 133–72. For an overview of issues surrounding the study of Roman sculpture produced in Asia Minor, see Ng, “Asia Minor.”

24) For an early fourth-century date, see Vermeule, “Two Masterpieces,” pp. 6, 9–10; Milkovich, Roman Portraits, p. 76, cat. 34; Hanfmann, Roman Art, p. 103, cat. 98; Vermeule, “Greek and Roman Portraits in North American Collections,” p. 103; Breckenridge, Likeness, p. 247, fig. 131; p. 248; Vermeule, Roman Imperial Art in Greece and Asia Minor, pp. 356, 364–65; Calza, Iconografia romana imperiale, pp. 268–69, cat. 181, pl. 94; Vostchinina, Le portrait romain, p. 194, cat. 79; Museum of Fine Arts, Boston, Romans and Barbarians, p. 110, cat. 118; Morgan, “Letter from Boston,” p. 499, fig. 8; Vermeule, Greek and Roman Sculpture in America, p. 372, no. 323. For a late fourth-century date (i.e., between 360 and 400), see Heintze, “Ein spätantikes Mädchenporträt,” pp. 78 (group 5, no. 4), 79, pls. 13d, 15a (A.D. 360/90); Breckenridge, “Head of a Lady” (A.D. 370/80); Ledig Heuser, “Head of a Female Figure,” p. 85 n. 1, cat. 27 (in comparison to a female portrait dated to A.D. 375/400); Manchester and Greuel, “Marble Portrait Head of a Young Woman” (listing previously assigned dates from the middle of the second to the late fourth century).

25) The hypothesis of a fourth-century revival of the second-century turban coiffure was first proposed by Tamara Uschakoff, who in 1928 dated a portrait of a woman in the collection of the State Hermitage Museum in Saint Petersburg, Russia (ГР-3581), to the first quarter of the fourth century. (As of this writing, this is still the date assigned to the portrait by the museum; see http://www.hermitagemuseum.org/wps/portal/hermitage/digital-collection/06.+Sculpture/220038/?lng.) Uschakoff’s dating was based in part on a comparison of the Hermitage portrait to certain coin portraits of Constantine’s mother, Helena; see Uschakoff, “Ein römisches Frauenporträt.” Subsequently other scholars dated numerous portraits with the turban hairstyle to late antiquity, spurring further debate; see Schade, Frauen in der Spätantike, pp. 98–99; Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, p. 66, cat. 86.

34) On the “Zopfkranzfrisur,” see Schade, Frauen in der Spätantike, pp. 96–98. On the portrait of Helena, see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, pp. 35–36, cat. 38, pls. 47–48. On the possibility that this portrait and four others depicting Helena were recarved from earlier portraits of empresses, including Faustina the Elder (A.D. 100–140/141) and Faustina the Younger (A.D. 125/130–175), see Prusac, From Face to Face, pp. 73, 156–57, cats. 482–86, pls. 139–40, figs. 151a–d, 152a–b. For a reconsideration of the hairstyle of the portrait of Helena in the Musei Capitolini, which might at one time have included features of the “Scheitzelzopffrisur” (parted braid hairstyle), specifically the locks over the nape of the neck that turn upward and are woven into braids covering the back of the head, see Arata, “La statua seduta.”

38) For the modern viewer, this is often taken to signify a strong interior spirituality or psychology on the part of the subject, which has frequently been connected to the rise of Christianity in late antiquity. On this point, see especially L’Orange, Studien zur Geschichte; Kitzinger, Byzantine Art in the Making. On the subjective nature of such psychological interpretations, see Prusac, From Face to Face, p. 60. More recently, the emphasis on the eyes in late antique portraits has been attributed instead to a symbolic claim to numerous positive qualities, such as vision, strength, knowledge, justice, energy, and personal intensity, depending on the broader context in which the image was displayed; see Smith, “Public Image of Licinius I,” pp. 198–201; Smith, “Late Antique Portraits in a Public Context,” pp. 182–89.

42) Kiilerich, Late Fourth Century Classicism, p. 117. Kiilerich noted that the facial features of the portrait resemble those of portraits dated to the Theodosian period. Klaus Fittschen and Paul Zanker similarly suggested that some portraits dated to late antiquity might have been created from earlier images; see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, p. 64, cat. 85. For an example, see Roman Woman in the Ny Carlsberg Glyptotek, Copenhagen (Hadrianic period, reworked in the 4th century; 710), in Johansen, Roman Portraits, vol. 3, pp. 196–97, cat. 87. For an example of a head (perhaps from an ideal statue) recarved into a female portrait with this hairstyle (Constantinian period; Museo Nazionale Romano, Rome, 61632), see Felletti Maj, I ritratti, pp. 157–58, cat. 316; Schade, Frauen in der Spätantike, pp. 172–73, cat. I 8, pl. 26. On the recarving of portraits, see especially Prusac, From Face to Face; and, most recently, Varner, “Reuse and Recarving.” For examples of portraits in the Art Institute’s collection bearing traces of ancient recarving, see cat. 5, Portrait Head of a Woman; cat. 15, Portrait Head of a Man; cat. 16, Portrait Head of a Youth; and cat. 19, Portrait Head of a Philosopher.

45) On the similarity of the turban coiffure to hairstyles of priestesses, see Calza, “Statua iconica femminile da Ostia,” p. 201; Harrison, “Constantinian Portrait,” p. 88. On the similarities to the headdresses or hairstyles of the Vestal Virgins, see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, p. 63, cat. 84; p. 66, cat. 86; Schade, Frauen in der Spätantike, p. 101. Additional studies of the portraits of the Vestal Virgins include Jordan, Der Tempel der Vesta, pp. 43–56, pls. 8–10; Van Deman, “Value of the Vestal Statues”; Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, pp. 92–94, cats. 134–35, pls. 160–61; Walker, “Portrait Head of a Life-Sized Statue of a Vestal”; Lindner, “Vestal Virgins and Their Imperial Patrons,” pp. 64–97; Lindner, Portraits of the Vestal Virgins.

Some Roman priestesses appear to have worn, rather than popular hairstyles, “nonfashion” coiffures with their hair hanging in long tresses or pulled back loosely and with locks hanging down in a manner resembling that of images depicting goddesses. For example, see the female head wearing a lunate diadem (identified here as a stephane), which might depict a priestess of Aphrodite, from the Hadrianic baths at Aphrodisias, in modern-day Turkey (2nd/3rd century A.D.; 66-270), in Dillon, “Female Head Wearing Stephanē.”

46) Lindner, Portraits of the Vestal Virgins, pp. 100–02 (on the infula), 164–86 (on the infula and turban coiffure); see also p. 174 for Lindner’s use of the term nonelite to refer specifically to women of equestrian rank and members of the middle class. John R. Clarke summarized the basic distinction between elite and nonelite members of Roman society as follows: “An elite Roman possessed the four prerequisites necessary to belong to the upper strata of society: money, important public appointments, social prestige, and membership in an ordo. (The ordines are those of senator, decurion, and equestrian.) The non-elite person lacked one or more of the four prerequisites.” Clarke, Art in the Lives of Ordinary Romans, p. 4. For an overview of the use of the modern term middle class in scholarship on ancient Roman society and culture, see Mayer, Ancient Middle Classes, pp. 1–21.

On this portrait head of a Vestal Virgin in the British Museum, London (c. 2nd century A.D.; 1979,1108.1), see Walker, “Portrait Head of a Life-Sized Statue of a Vestal”; see also Lindner, Portraits of the Vestal Virgins, pp. 130–32, cat. 2, figs. 26–27.

53) For an overview of the public roles of women in the Latin West, including the roles of priestess and benefactress, see Hemelrijk, “Public Roles for Women.” For the public prominence of imperial cult priestesses in the Latin West, see Hemelrijk, “Local Empresses,” p. 343. On women’s public roles in the Greek East, see MacMullen, “Women in Public”; Bremen, Limits of Participation. For the possibility that the turban coiffure was worn by priestesses to visually evoke the infula of the Vestals, see Lindner, Portraits of the Vestal Virgins, p. 182. For an example of a public honorific statue from a religious sanctuary that depicts a priestess wearing the turban coiffure, see the portrait of Minia Procula, who was named flaminica perpetua (public priestess appointed for life), from Bulla Regia in modern-day Tunisia, now in the Musée du Bardo in Tunis (2nd century A.D.; C 1020). For this portrait, see Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, p. 63, cat. 84, item l; Lindner, Portraits of the Vestal Virgins, p. 182.

55) On the palla, see Olson, Dress and the Roman Woman, pp. 33–36; Cleland, Davies, and Llewellyn-Jones, Greek and Roman Dress, pp. 136–37. The veil was also worn by brides and matrons and thus was associated with “wifely virtues”; see Bartman, Portraits of Livia, pp. 44–45. See also Sebesta, “Symbolism in the Costume of the Roman Woman,” pp. 48–49. On the rarity of depictions of women burning incense, see Lindner, Portraits of the Vestal Virgins, pp. 244–45.

Additional portraits depicting veiled women wearing the turban coiffure (although not necessarily identified as priestesses) include the following: (1) statue of a woman, Small Herculaneum Woman type (early Hadrianic period; Musei Capitolini, Palazzo Braschi, Rome), in Fittschen and Zanker, Katalog der römischen Porträts, vol. 3, pp. 63–65, cat. 85, pl. 107; (2) statue of a woman, Large Herculaneum Woman type, from Philadelphia (modern-day Alaşehir, Turkey) (Trajanic or Hadrianic period; Manisa Müzesi, Turkey, 1298), in Inan and Rosenbaum, Roman and Early Byzantine Portrait Sculpture, p. 163, cat. 214, pls. 115, 2, and 118, 1–2; (3) statue of a woman, Pudicitia type, from Ostia (Antonine period or 3rd century; Museo Ostiense, Ostia, Italy, 22), in Calza, Iconografia romana imperiale, pp. 251–52, cat. 165, pls. 87–88, figs. 307–10. Kathrin Schade has suggested that popular female statuary types such as those noted above, each of which depicts the figure standing in a closed, seemingly restrained pose, might have conveyed a message of feminine virtue, much like the turban coiffure; see Schade, Frauen in der Spätantike, p. 102. However, Jennifer Trimble has argued that the self-contained gestures of such statuary types “do not seem to be a specifically feminine characteristic, connoting, for example, modesty,” given the use of closed poses in male portraiture; see Trimble, Women and Visual Replication, p. 165. For an overview of the main statuary types used in portraits of women, which include the three types mentioned here, see Fejfer, Roman Portraits in Context, pp. 335–44.

59) The thin-section petrography and X-ray diffraction were performed at the Laboratorio di Analisi dei Materiali Antichi (LAMA), Università IUAV, Venice, by Lorenzo Lazzarini, Director, LAMA. The purpose of the minero-petrographic analysis was to determine the following parameters: (1) the type of fabric (either homeoblastic or heteroblastic, the former consisting of grains of roughly equal dimensions and the latter consisting of grains of variable dimensions), which is directly related to the type of metamorphism (equilibrium, nonequilibrium, retrograde, or polymetamorphism); (2) the boundary shapes of the calcite/dolomite grains, which are also correlated with the types of metamorphic events that generated the marble; (3) the maximum grain size, a parameter that has diagnostic significance because it is linked to the grade of metamorphism achieved by the marble; and (4) the qualitative and semiqualitative presence of accessory minerals, which are sometimes of diagnostic value. In the formulation of the petrographic description, previous specific studies of the most important “major” ancient marbles (Gorgoni et al., “Reference Database”; Lazzarini, Moschini, and Stievano, “Contribution to the Identification”), other archaeometric studies of “minor” marbles, and seminal treatises on petrotectonics (e.g., Spry, Metamorphic Textures) were taken into consideration. Lazzarini, “Report on the Results,” pp. 1–2. This essay is greatly indebted not only to Lorenzo Lazzarini’s analysis, which facilitated a better understanding of this object in the newly cast light of its secure attribution with respect to marble type and provenance, but also to his graciousness and generosity in sharing the vast knowledge and wisdom acquired over the course of his decades-long career.

63) A certain amount of marble from Dokimeion seems to have been exported to imperial Rome, especially for statuary, because of its very fine grain and good carving properties. Marmor docimenum was already used in Hellenistic times (for example, at Tralles); it continued to be exploited in Roman and Byzantine times, for statuary as well as sarcophagi (for example, of the Sidamara type), and is still quarried today. The quarries were probably under the control of the nearby ancient village of Dokimeion, in the province of Synnada in ancient Phrygia, hence its other ancient names, marmor synnadicum and marmor phrygium. Lazzarini, “Report on the Results,” p. 6. For further reference, see Monna and Pensabene, Marmi dell’Asia Minore, pp. 29–77; and Pensabene, “Cave de marmo bianco.” For information on the quarries at Dokimeion, see Fant, Cavum antrum Phrygiae, pp. 17–320.

64) The stable isotope ratio analysis was done at the Isotopen Labo, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Germany, and interpreted by Lorenzo Lazzarini, Director, LAMA. Isotopic characterization has proved to be very useful in the identification of ancient marble. Its use is becoming more and more widespread owing to its outstanding sensitivity, the small quantity of material necessary for the analysis, and the availability of a rapidly growing isotopic database, often associated with other laboratory methodologies (Attanasio, Brilli, and Ogle, Isotopic Signature of Classical Marbles; Barbin et al., “Cathodoluminescence”; Gorgoni et al., “Reference Database”; Lazzarini, “Archaeometric Aspects of White and Coloured Marbles”), that permits increasingly trustworthy comparisons, especially if the isotopic data are evaluated together with the minero-petrographic results from the same samples, as in the present study. The present study used reference data from petrographic databases as well as data compiled from thin-section samples taken in fifty ancient Mediterranean marble quarries. Lazzarini, “Report on the Results,” pp. 2, 3.

Isotopic analysis was carried out on the carbon dioxide derived from a small portion (20–30 mg) of the powdered sample obtained through dissolution by phosphoric acid at 25°C in a vacuum line, according to the procedure suggested by McCrea, “Isotopic Chemistry of Carbonates,” and Craig, “Isotopic Standards for Carbon and Oxygen.” The resulting carbon dioxide was then analyzed using mass spectrometry. The mass spectrometer is fitted with a triple collector and permits the measurement of both isotopic ratios (¹³C/¹²C and ¹⁸O/¹⁶O) at the same time. The analytical results are conventionally expressed in δ units, in parts per thousand, in which δ = (R_sample/R_std – 1) × 1000, and R_sample and R_std represent the isotopic ratios of oxygen and carbon in the sample and in the reference standard, respectively. The standard adopted is PDB, or δ¹³C, for both oxygen and carbon. The PDB standard for isotope ratio measurements in carbonates is the rostrum of the Belemnitella americana, a Cretaceous marine fossil from the Pee Dee Formation in South Carolina. Lazzarini, “Report on the Results,” p. 2.

66) The topic of joining heads and appendages onto marble sculpture, during either initial fabrication or subsequent alteration, is a rich and complex one. For a thorough explanation of the ways in which heads and other appendages were affixed either by pinning or by the use of sockets and tenons, as well as the chronology of these methods, see Claridge, “Ancient Techniques of Making Joins.” Diagrammatic representations of the many shapes and profiles of tenons commonly found on ancient marble statuary can be found on page 145 of Claridge’s text. For further information, see Claridge, “Marble Carving Techniques.”There is a very remote possibility that the base of the neck could have terminated in a flat surface. Were this the case, the object would most likely have been a replacement head. For further information on the technical aspects of removing, fitting, and securing replacement or repurposed heads, many of which can be identified by flat breaks through the neck and keying across the break surfaces, see Claridge, “Late Roman Portrait Head.” For a broad overview of the reuse and replacement of portrait heads, see Varner, “Reuse and Recarving.”

67) The so-called honeycombing method, in which a groove was cut by boring a row of holes at short intervals and then clearing the remaining stone out with a chisel or channeling tool, is the most laborious means by which to create grooves in marble. Sculptors may have chosen this method because it was deemed safer and offered more control. In another method, the running drill, the drill was rotated at a sharp angle to the stone, cutting a groove as it was moved along. The same tool was used for both; only the method of application differed. Great controversy surrounds the running drill with respect to the date of its introduction and the extent of its use. Olga Palagia discussed the physical evidence of the drill marks left in the Parthenon pediments (where the heart of the controversy lies) and found sufficient proof that their sculptors employed both methods. Palagia, “Marble Carving Techniques,” 253–60. Amanda Claridge, on the contrary, has stated that the running drill is “a phantom, based on a misunderstanding of the German ‘laufende Bohrer.’” Claridge, “Marble Carving Techniques,” p. 108.

69) “Emery, corundum, volcanic scoria, and quartz sand are variously attested.” Claridge, “Marble Carving Techniques,” p. 108. Anecdotal evidence suggests that the Romans may also have used “rosin or colophon wax, the wax athletes use on their hands. They rubbed it on a very soft, thin piece of lead which had been wrapped around a piece of cloth, giving rosin-coated lead with a cushion inside.” A later account that may have roots in antiquity comes from the sculptor Carradori, who, “in his 1802 instruction manual for sculpture, gives a formula for polishing: break up the bones of a goat, cook them to a charcoal and put the charcoal in a mortar and pestle. Grind the mixture to a fine powder, mix it with water and then rub the stone to get a gloss polish.” Modern sculptors apply oxalic acid in powder form to a piece of burlap dampened with water and rub the sculpture to a high final polish. Rockwell, Rosenfeld, and Hanley, “Compleat Marble Sleuth,” p. 31. Ancient sculptors might also have used acids as a finishing technique.

71) Ultraviolet-induced visible light fluorescence was among the first technical examination methods to be employed in the service of investigating art objects, and, although there is some subjectivity in the interpretation of the fluorescence, it remains the technique most universally chosen as a first step in examining works of art. UV radiation causes unique patterns of fluorescence in a wide array of materials, and the color and intensity of the fluorescence help to differentiate between original materials and alterations, be they wholesale additions or surface modifications. Very broadly speaking, with respect to marble, a newly broken or modern surface will appear bright purple when exposed to UV radiation, whereas an intact, ancient surface will appear white, yellow, or amber. An older but not necessarily ancient surface will appear more blue. See Rorimer, Ultra-Violet Rays, pp. 17–23. This phenomenon can be exploited to reveal parts of a sculpture that may have been recently cleaned or treated; may have been recut or replaced; or may retain pigments, coatings, or colorants. See Grossman and Maish, “Authenticity of a Classical Attic Funerary Monument.” For further information on the refinement of UV/visible fluorescence imaging since 1931 and its reliability in the examination of marble artifacts, see Polikreti, “Detection of Ancient Marble Forgery”; Polikreti and Christofides, “Composition Profiles and Time Evolution”; Polikreti and Christofides, “Laser-Induced Micro-photoluminescence”; and Polikreti and Christofides, “Role of Humic Substances.”