The Vanishing Black Indian: Revisiting Craniometry and Historic Collections

Pamela L. Geller

Objectives: This article uses craniometric allocation as a platform for discussing the legacy of Samuel G. Morton's collection of crania, the process of racialization, and the value of contextualized biohistoric research perspectives in biological anthropology. Materials and Methods: Standard craniometric measurements were recorded for seven Seminoles in the Samuel G. Morton Crania Collection and 10 European soldiers from the Fort St. Marks Military Cemetery; all individuals were men and died in Florida during the 19th century. Fordisc 3.1 was used to assess craniometric affinity with respect to three samples: the Forensic Data Bank, Howells data set, and an archival sample that best fits the target populations collected from 19th century Florida. Discriminant function analyses were used to evaluate how allocations change across the three comparative databases, which roughly reflect a temporal sequence. Results: Most Seminoles allocated as Native American, while most soldiers allocated as Euro-American. Allocation of Seminole crania, however, was unstable across analysis runs with more individuals identifying as African Americans when compared to the Howells and Forensic Data Bank. To the contrary, most of the soldiers produced consistent allocations across analyses. Repeatability for the St. Marks sample was lower when using the archival sample database, contrary to expectations. For the Seminole crania, Cohen's κ indicates significantly lower repeatability. A possible Black Seminole individual was identified in the Morton Collection. Discussion: Recent articles discussing the merits and weaknesses of comparative craniometry focus on methodological issues. In our biohistoric approach, we use the patterning of craniometric allocations across databases as a platform for discussing social race and its development during the 19th century, a process known as racialization. Here we propose that differences in repeatability for the Seminoles and Euro-American soldiers reflect this process and transformation of racialized identities during 19th century U.S. nation-building. In particular, notions of whiteness were and remain tightly controlled, while other racial categorizations were affected by legal, social, and political contexts that resulted in hybridity in lieu of boundedness.

Received: 27 May 2016 | Revised: 26 September 2016 | Accepted: 30 September 2016 DOI 10.1002/ajpa.23115 RESEARCH ARTICLE The vanishing Black Indian: Revisiting craniometry and historic collections Pamela L. Geller1 | Christopher M. Stojanowksi2,3 1 Department of Anthropology, University of Miami, Miami, Florida 2 School of Human Evolution and Social Change, Arizona State University, Arizona Abstract Objectives: This article uses craniometric allocation as a platform for discussing the legacy of Samuel G. Morton’s collection of crania, the process of racialization, and the value of contextual- 3 Center for Bioarchaeological Research, Arizona State University, Arizona Correspondence Pamela L. Geller, Department of Anthropology, University of Miami, P.O. Box 248106, Coral Gables, Miami, FL 33124. Email: [email protected] Funding Information This research was supported by the University of Miami’s Provost Research Award (2013 and 2014). ized biohistoric research perspectives in biological anthropology. Materials and Methods: Standard craniometric measurements were recorded for seven Seminoles in the Samuel G. Morton Crania Collection and 10 European soldiers from the Fort St. Marks Military Cemetery; all individuals were men and died in Florida during the 19th century. Fordisc 3.1 was used to assess craniometric aﬃnity with respect to three samples: the Forensic Data Bank, Howells data set, and an archival sample that best ﬁts the target populations collected from 19th century Florida. Discriminant function analyses were used to evaluate how allocations change across the three comparative databases, which roughly reﬂect a temporal sequence. Results: Most Seminoles allocated as Native American, while most soldiers allocated as EuroAmerican. Allocation of Seminole crania, however, was unstable across analysis runs with more individuals identifying as African Americans when compared to the Howells and Forensic Data Bank. To the contrary, most of the soldiers produced consistent allocations across analyses. Repeatability for the St. Marks sample was lower when using the archival sample database, contrary to expectations. For the Seminole crania, Cohen’s j indicates signiﬁcantly lower repeatability. A possible Black Seminole individual was identiﬁed in the Morton Collection. Discussion: Recent articles discussing the merits and weaknesses of comparative craniometry focus on methodological issues. In our biohistoric approach, we use the patterning of craniometric allocations across databases as a platform for discussing social race and its development during the 19th century, a process known as racialization. Here we propose that diﬀerences in repeatability for the Seminoles and Euro-American soldiers reﬂect this process and transformation of racialized identities during 19th century U.S. nation-building. In particular, notions of whiteness were and remain tightly controlled, while other racial categorizations were aﬀected by legal, social, and political contexts that resulted in hybridity in lieu of boundedness. KEYWORDS biohistory, craniometry, history of physical anthropology, racialization, repatriation 1 | INTRODUCTION and nation-building during the 19th century. U.S. Army surgeons, for Samuel G. Morton (b. 1799, d. 1851) is a disciplinary ancestor who has tleﬁelds. In doing so, they set a precedent that guided future relations long engendered controversy. He is perhaps best known for the size- between science, government, and the military. Physical anthropology able collection of crania, 967 in total, he acquired prior to his death. continues to navigate the intellectual and ethical repercussions of these Less well known, however, is the means by which he acquired them, a historic events (Cantwell, 2000; Geller, 2015; Kakaliouras, 2008, 2012; process often couched within a violent history of settler colonialism Mukhopadhyay & Moses, 1997; Ousley & Jantz, 2005; Roseman, instance, often gifted him with “specimens” taken from graves and bat- Am. J. Phys. Anthropol. 2016; 1–18 wileyonlinelibrary.com/journal/ajpa C 2016 Wiley Periodicals, Inc. V | 1 2 | GELLER AND STOJANOWKSI 2014; Rose, Green, & Green, 1996). From his study of these crania, Hefner, 2009; Hefner & Ousley, 2014; Hefner, Spradley, & Anderson, Morton developed craniometry, an analytical method that was founda- 2014; Klales & Kenyhercz, 2015; Spradley & Jantz, 2016), despite the tional for physical anthropology in the U.S. and for which he gained controversy its assessment generates (Brace, 1995; Albanese & Sand- international prominence. Based on calculations of cranial capacity, ers, 2006; Armelagos & Van Gerven, 2003; Cartales, 2015). An under- which was assumed to measure intelligence, Morton (1839) made standing of human diﬀerence in typological terms has shifted to a inferences about racial hierarchical ranking. Big brained Caucasians, he populational view, however. Determination of “race” (or ancestry) is argued, were superior to small skulled Native Americans and people of more accurate, investigators maintain, when multiple measurements African descent. His work also contributed to formation of the “Ameri- are taken; the standards established by Howells (1973, 1989, 1995) can School” of ethnology; along with Louis Agassiz, George Gliddon, are particularly inﬂuential in research design and implementation. For and Josiah Nott, Morton advanced polygenism, the misguided notion example, the variable deﬁnitions provided by Howells are still in wide- that human races have distinct origins (Gould, 1981; Horsman, 1975; spread usage and the 28 populations comprising the Howells data set Stanton, 1960). Euro-Americans invested in colonial expansionism and provide a reference sample for both CranID (Wright, 2012) and Fordisc enslavement cited Morton’s and colleagues’ studies to legitimize state (Ousley & Jantz, 2005), computer programs that determine ancestry sanctioned violence against races they deemed separate and intellectu- through discriminant function analysis. Their creators argue that a cor- ally inferior. The scrutiny of this history continues today. Accounts of relation does exist between cranial morphology and geographic origin, “skull wars” and “bone rooms,” which have been written for a more and given this relationship researchers can assess population aﬃnity general audience, challenge the negotiation of our discipline’s past and when they use the programs with correct reference samples and follow future within the public realm (see Fabian, 2010; Redman, 2016; proper statistical protocol (Fordisc 3.1 has an 83 page user manual) Thomas, 2000). (see also Freid, Spradley, Jantz, & Ousley, 2005; Ousley et al., 2009; Given the racist underpinnings of his work, anthropologists have Wright, 1992, 2008). come to regard craniometry with ambivalence. Franz Boas’s (1912) Many evaluative studies of forensic discriminant function analysis immigrant study with its focus on the cephalic index provided a turning exist in the literature (Belcher, Williams, & Armelagos, 2002; Campbell point [and still engenders debate (Gravlee, Bernard, & Leonard, 2003a, & Armelagos, 2007; Cross & Wright, 2015; Elliott & Collard, 2009; Gravlee, Bernard, & Leonard 2003b; Jantz & Logan, 2010; Sparks & Freid et al., 2005; Kallenberg & Pilbrow, 2012; Keita, 2007; Kosiba, Jantz, 2002, 2003; Sutphin, Ross, & Jantz, 2014)]. Evidence for cranial 2000; L’Abbe, Kenyhercz, Stull, Keough, & Nawrocki, 2013; Leathers, plasticity pointed to environmental inﬂuences on cranial form. The rela- Edwards, & Armelagos, 2002; Ross, 1999; Ross, Slice, Ubelaker, & tionship between genotype and phenotype was also brought into ques- Falsetti, 2004; Sejrsen, Lynnerup, & Hejmadi, 2005; Sierp & Henne- tion, as was race’s immutability and discreteness. Scholars have berg, 2015; Spradley & Jantz, 2016; Williams, Belcher, & Armelagos, continued to oﬀer critical assessment of Morton’s conclusions. Stephen 2005; Wittwer-Backofen et al., 2014; Ubelaker, Ross, & Graver, 2002; Jay Gould (1978, 1981), who never directly examined the crania, was Urbanova, Ross, Jurda, Nogueira, 2014). Detractors argue that these the most notable critic of their scientiﬁc value. Racism, he and others programs are of little use if unknown individuals do not fall within the agree, has had a heavy hand in making and naturalizing race (see also reference populations’ geographic or temporal limits (Belcher et al., Barkan, 1992; Blakey, 1987; Brace, 2005; Montagu, 1997; Stocking, 2002; Campbell & Armelagos, 2007; Elliott & Collard, 2009; Keita, 1982; Thomas, 2000). Among anthropologists, the crania collection still 2007; Kosiba, 2000; Leathers et al., 2002; Williams et al., 2005). The generates research queries about methodology and scientiﬁc racism Forensic Data Bank (FDB), for example, is appropriate for forensic (e. g., Lewis et al., 2011; Lieberman, 2001). Lewis et al. (2011), for anthropologists with contemporary medicolegal concerns (Dirkmaat, instance, recently re-measured 308 crania and found Morton’s scien- Cabo, Ousley, & Symes, 2008; Jantz, 2001; Jantz & Meadows Jantz, tiﬁc methods “sound,” a point that Michael (1988) drew attention to 2000; Ousley & Jantz, 1998, 2012; Wescott & Jantz, 2005). Yet, the some two decades earlier. Such technical accuracy, Lewis et al. claim, database may be less applicable to remains that are not modern in demonstrates that Morton’s results were not racially biased. Other date. Additionally, reliability can be reduced by using too few (Hubbe & scholars have deemed this conclusion illogical or at least in need of Neves, 2007) or too many measurements (Konigsberg, Algee-Hewitt, & debate (Marks, 2011; Tattersall, 2013; Weisberg, 2014). Steadman, 2009; Ousley et al., 2009; Ubelaker et al., 2002) and also by Despite the controversy, and to varying degrees, forensic anthro- loading the allocation space with samples that are not of likely rele- pologists continue to voice support for craniometry though eschew the vance to the speciﬁc research question or case at hand (Algee-Hewitt, hierarchical ranking of races (Bass, 2005; Byers, 2010; Gill, 1998; Ous- 2016; Konigsberg et al., 2009). That is, human cranial variation is not ley, Jantz, & Freid, 2009; Relethford, 1994, 2009; Sauer, 1992, 1993; discretely distributed (Strauss & Hubbe, 2010), although it is broadly Stojanowski & Duncan, 2009). Perhaps, they do so because of the and geographically patterned (Manica, Amos, Balloux, & Hanihara, method’s connection to ancestry assessment, which the subﬁeld’s prac- 2007; Relethford, 1994, 2004, 2009; Roseman & Weaver, 2004; von titioners have long regarded as a key aspect of the biological proﬁle Cramon-Taubadel, 2014), and to assume regional discreteness asks too (Birkby, 1966; Gill, 1998; Gill & Rhine, 1990; Giles & Elliot, 1962; much of the data. Critics also echo Boas’s statements about plasticity Sauer, 1992). Work on the racial dimensions of biological proﬁles, now in their rejoinder that “racial” traits are not unchanging through time commonly referred to as ancestry, continues (Berg & Ta’ala, 2015; (Belcher et al., 2002; Williams et al., 2005), which further emphasizes GELLER AND T A B LE 1 | STOJANOWKSI 3 Sample of Seminole with information about collector, age, sex, identiﬁcation, and provenience Current catalogue # Collector Age Sex Identiﬁcation Provenience 97-606-604 Dimick, Justin 401 M Warrior Battle of St. Joseph’s Plantation, St. Johns Co., FL 97-606-698 Abert, John J. 30–40 M Warrior FL 97-606-707 Abadie, Eugene H. 20–30 M? Eoklo-Emathla 12 miles south of Sewanee River (near Tampa), FL 97-606-730 Abadie, Eugene H. 25–35 M Warrior Lake Okeechobee Battleﬁeld, FL 97-606-732 Abadie, Eugene H. 30–45 M Warrior Lake Okeechobee Battleﬁeld, FL 97-606-754 Walker, Joseph 30–45 M Warrior FL 97-606-1105 Robertson, Francis M. 35–45 M? Warrior Dade’s Battleﬁeld, Sumter Co., FL the importance of contextualization when informing a speciﬁc analysis essarily a place of origin, for those men whose lives converged as a (Algee-Hewitt, 2016; Brues, 1992; Konigsberg et al., 2009; Stojanowski consequence of historical events and political circumstances. & Duncan, 2009). Here, we present a series of craniometric allocations using the For- This article contributes further to the ongoing debate surrounding disc 3.1 software. In addition to using the Howells data set and the For- craniometric analysis by reexamining crania in the Morton Collection. ensic Data Bank, we also implement the speciﬁc population approach Speciﬁcally, we focus on those individuals acquired from Florida that of Brues (1992), which uses samples of the most appropriate temporal Morton labeled Seminole (n 5 16).1 Although, as noted above, there is and geographic provenance for comparative purposes. While not a cor- considerable controversy surrounding craniometry, we do not oﬀer rective for all of the challenges of craniometry, implementing a contex- another assessment of whether the technique works or not. Nor are tualized approach to ancestry assessment does allay concerns about we concerned with evaluating the protocols needed to ensure its accu- overly typological approaches wherein racial taxonomies allow substitu- racy. Instead, we adopt a best practices methodology (following Ousley tion of samples across broad swaths of space and time. Furthermore, & Jantz, 2005, 2012) to demonstrate how study of the Morton Collec- we show that the inferences drawn from biological data deepen and tion extends understanding of population aﬃnity and social concep- become relevant to a broader array of scholarly and nonscholarly audi- tions of race by adopting a biohistorical and contextual approach (for ences when researchers engage with historically important collections example, de la Cova, 2010, 2011, 2012; Edgar, 2009; Rankin-Hill, and archival sources. There is much information to extract from these 1997; Rose, 1985). To be clear, the focus here is not on race per se. resources that enriches biological anthropology beyond simple ques- Rather, the concern is with racialization. We see the race concept as tions of practice and method (de la Cova, 2010, 2011, 2012; Herring & being constrained to identities, while racialization requires that we Swedlund, 2003; Watkins, 2012; Wittwer-Backofen et al., 2014). For think about the sociopolitical and historical processes at work in the example, the research presented here was undertaken at nine institu- constitution (and reconstitution, for they are not static) of such social tions (the University of Pennsylvania Museum of Archaeology and identities. In doing so, we use the collection to explore shifting notions of hybridity and race in the United States over the last two centuries. More speciﬁcally, in this article we analyze the crania of seven male Seminole “warriors” in the Morton Collection, supplemented with the nearly contemporaneous crania of 19th century U.S. Army soldiers (all presumably Euro-American) buried in Florida’s Fort St. Marks Military Cemetery. These two collections are not combined haphazardly. Seminole Indians and the U.S. military clashed throughout Florida from roughly 1812 to 1858, a period of time that is more familiarly known as the Seminole Wars. The men included in this analysis died during the course of these wars and, situated as such, were then physically buried in Florida. Florida, then, represents a spatial backdrop, not nec- Anthropology; University of Miami’s Special Collections; American Philosophical Society; Academy of Natural Sciences; Historical Society of Pennsylvania; Library Company; College of Physicians of Philadelphia; Wistar Institute; and National Archives). Source materials from these archives included personal letters, photographs, original research notes, and book marginalia that held information about Morton’s biography, the collection’s history, individuals who obtained crania for him, and named decedents in the collection. Our discussion, then, involves both forensic and bioarchaeological frames to “people the past” in a way that humanizes biophysical data, demonstrates the importance of the preservation and re-analysis of collections, and adds new perspectives on craniometry that embrace analytical complexities as a space for exploring the process of racialization within the U.S. 1 For a comparable reassessment of the Morton Collection, see Emily Renschler’s (2007) study of crania belonging to enslaved African-born decedents who died shortly after arriving in Cuba (n 5 51). Archival documentation contains information about the geographic origins of these decedents, which craniometric analysis in turn supports. Renschler did, however, ﬁnd a high degree of individual heterogeneity within the sample. In explanation, she notes, historical accounts of the Cuban slave trade indicate native-born Africans were from diverse cultures and locations. 2 | MATERIALS AND METHODS 2.1 | Materials 2.1.1 | The Samuel G. Morton crania collection The materials used in this study derive from the Samuel G. Morton Crania Collection, which is housed at the University of Pennsylvania 4 | GELLER AND STOJANOWKSI Museum of Archaeology and Anthropology. In total, Morton acquired roughly contemporaneous with crania in the Morton Collection. It is 967 skulls prior to his death in 1851. Approximately 25% of the Mor- important to note that in both analyses we excluded populations that ton Collection, about 250 to 300 individuals, was comprised of Native are not relevant to the speciﬁc research question (Konigsberg et al., American decedents with diverse cultural aﬃliations. Here we are con- 2009; Ousley & Jantz, 2012). cerned with crania labeled Seminole, an attribution supported by Finally, we compared each specimen to a broader array of samples archival sources (Geller, 2015). That is, Morton rarely acquired crania that are contemporary with or slightly older than the Seminole and St. ﬁrsthand and much of the collection was obtained because he was a Marks individuals (some of which may have been born in the late 18th proliﬁc letter writer with an extensive network of colleagues willing to century). Given the changes in biology that occurred throughout the procure “specimens” for him. Military oﬃcers answered many of his colonial period due to demographic collapse, in-migration, and gene ﬂow, requests. Men living or stationed in Florida, for instance, acquired six- this “Contemporary” database best reﬂects implementation of the spe- teen crania (some of which had mandibles) between 1831 and 1852. ciﬁc population approach advocated by Brues (1992). For the sake of This period of time roughly coincides with the Second Seminole War consistency, and to reﬂect 19th century notions of human variation, (1835–1842). Hence, these individuals’ crania were acquired in the these individual sites were collapsed into three populations: Euro- aftermath of violent conﬂicts and forced relocation policies associated American, African American, and Native American. Sample details are as with U.S. expansionism (Geller, 2015). While all Seminoles were follows. impacted by these events, here we focus on the seven crania of adult The Euro-American sample (n 5 85 males) represents constituent males (or possible males). Archival documents indicate that these indi- English, Scottish, Spanish, and French populations that migrated to the viduals were warriors, and in many cases military oﬃcers collected their New World and comprised the early “White” population of the east crania directly from battleﬁelds (see Table 1). coast. This sample is most appropriate because archival data suggests 2.1.2 | The Fort St. Marks military cemetery the St. Marks soldiers predominantly originated from the mid-Atlantic The Fort St. Marks Military Cemetery (8WA108), hereafter the St. Marks cemetery, was excavated in 1966 and is currently housed at Florida State University. The sample includes 19 male soldiers who died between 1818 and 1819 while stationed at the fort, which is located in Wakulla County, Florida. They were likely associated with Andrew Jackson’s invasion of Florida during the First Seminole War. All were assumed to be of European descent based on archival records (Dailey, Morell, & Cockrell, 1972). The site oﬀers a useful point of comparison to the Seminole crania in the Morton Collection; both samples date to the ﬁrst half of the 19th century and derive from Florida contexts. region of North Carolina and Virginia (Dailey et al., 1972), which matches the sample composition of the J. Lawrence Angel archives at the Smithsonian Institution (see Angel, 1976). The Angel archives data consist of 17th, 18th, and 19th century individuals predominantly from the eastern U.S., including historic samples from well-known colonial sites such as Carter’s Grove, Flowerdew Hundred, Gloucester Point, Martin’s Hundred, and St. Anne’s Church, among other singleton specimens from a variety of contexts throughout the U.S. The African American sample (n 5 149 males) also derives from the Angel archives and includes data from Catoctin Furnace and College Landing, among other proveniences heavily representing Mary- 2.1.3 | Comparative databases land, Virginia, and Pennsylvania. These data were supplemented with Seminoles in the Morton Collection and St. Marks cemetery soldiers specimens from the Oakland Cemetery, Atlanta (Beck, 1980), the Virgin were compared to three databases. We ﬁrst compared individuals to the Islands (Buxton, Trevor, & Julien, 1938), and Cunningham Mound D, Forensic Data Bank with the following populations selected: White Male Georgia (Thomas, South, & Larsen, 1977). (20th century birthdate, national distribution), Black Male (20th century The Native American sample (n 5 138 males) includes both postcon- birthdate, mostly mid-Atlantic, largely from the Terry Collection; see tact samples from the Howells database (Arikara and Santa Cruz) supple- Hunt & Albanese (2005)), and Amerind Male (Ousley & Jantz, 2005). The mented with data from late precontact (Irene Mound, ca. 1300–1500) Amerind male sample includes a variety of proveniences including some and early postcontact (Santa Catalina de Guale (ca. 1600–1675), Santa mid- to late-19th century individuals (Ousley & Jantz, 2005). Catalina de Guale de Santa Maria (ca. 1675–1700) sites located along We then compared each cranium to the Howells database in For- the Georgia coast. The colonial period sites from Georgia are the abso- disc with the following male populations selected: 19th century Black, lute best geographic ﬁt for the Florida Seminole. They are, however, at 19th century White, Arikara, and Santa Cruz. The Black and White least a century older than the 19th century Seminoles in the Morton Col- sample data derive from the Hamman-Todd and Terry collections and lection. As such, they represent southeastern U.S. Native American biol- are more relevant to this analysis than the Old World European (Norse ogy prior to tribal ethnogenesis (Stojanowski, 2005, 2013). This process (Medieval Norway), Zalavar (9th to 11th century Hungary), Berg (multi- was set in motion with the destruction of the Spanish missions during generation family, Austria) or African (Dogon (Mali), Teita (Kenya)) sam- the ﬁrst decade of the 18th century, and ultimately gave rise to the ples included in the Howells database (according to Bateman (2002) Creek, Seminole, and Mikasuki identities recognized today. the Kongo-Angola region was the source of many Black Seminole pop- In combination, the Contemporary, Howells, and Forensic Data ulations; see also Humphries (2011)). While neither the Santa Cruz Bank databases represent a loose temporal sequence with some over- (“recent” Chumash, California) nor Arikara (17th century, South Dakota) lap of specimens deriving from the inclusion of data in the Angel samples are a strong geographic ﬁt for the Seminole, both samples are archives. This sequence allows us to evaluate how craniometric GELLER AND | STOJANOWKSI allocation for each specimen changes through time, where time represents both emergent processes of contested biohistory and hybridity 5 Measurement abbreviations and deﬁnitions after Howells (1989) and Martin (1928) T A B LE 2 as well as secular changes of a strictly biological nature (see Albanese, Abbreviation Deﬁnition GOL Glabello-occipital length XCB Maximum cranial breadth ZYB Bizygomatic breadth 2.2 | Methods BBH Basion-bregma height Morton’s sex assessments were conﬁrmed based on cranial variation BNL Basion-nasion length following Buikstra & Ubelaker (1994). Tooth wear (Miles, 2001) and BPL Basion-prosthion length ectocranial suture closure (Meindl & Lovejoy, 1985) were used to esti- MAB Palate breadth, external MAL Maxillo-alveolar length the protocol of Howells (1989) and Martin (1928). Data from the St. AUB Biauricular breadth Marks cemetery were recorded by CMS using the same measurement NPH Nasion-prosthion height and observational protocols as above (see Table 2). UFHT Upper facial heighta WFB Minimum frontal breadtha ples using Fordisc 3.1. Initial analysis runs used all available measure- UPBR Upper facial breadtha ments, but this usually resulted in an over-ﬁt model reﬂected in low v2 NLH Nasal height and rank order typicality probabilities (Ousley & Jantz, 2005, 2012). To NLB Nasal breadth OBB Orbit breadth, left OBH Orbit height, left (forward mean %, forward min %, forward Wilks) and present the EKB Biorbital breadth results that maximized the cross validation rate, typicality probabilities, DKB Inteorbital breadth FRC Nasion-bregma/frontal chord PAC Bregma-lambda/parietal chord provided in the extended results option in Fordisc 3.1 (Ousley & Jantz, OCC Lambda-opisthion/occipital chord 2005, 2012). FOL Foramen magnum length FOB Foramen magnum breadth MDH Mastoid height/length most spatially and temporally appropriate samples available. All data ASB Biasterionic breadth from published and archival sources were entered into a database, and ZMB Bimaxillary breadth then variables and individuals were winnowed until an acceptable level MOW Mid-orbitale width JUB Bijugal breadth OBH, OBB, EKB, NLH, NPH (see Table 2)), which is generally consid- NOL Nasio-occipital length ered suﬃcient for craniometric allocation depending on the degree of WMH Cheek height between group dispersion (Ousley & Jantz, 2005, 2012). Case-wise XFB Maximum frontal breadth Osley, & Tuck, 2012; Angel, 1976; Ousley & Jantz, 1998; Jantz, 2001; Jantz & Meadows Jantz, 2000; Ross, Ubelaker, & Kimmerle, 2011; Wescott & Jantz, 2005). mate age. Seminole individuals in the Morton Collection were measured by PLG using traditional landmark-based measurements following The population aﬃnity of each specimen was assessed with respect to the Contemporary, Howells, and Forensic Data Bank sam- correct this, we used the stepwise procedure where the maximum number of variables allowed in the model was equal to one-third of the smallest comparative sample size. We ran all three stepwise options and number of variables used. In most cases, the allocation results were consistent regardless of selection method. Homogeneity of variance-covariance matrices was evaluated using the Kullback test The Contemporary population database derived from archival sources was used to allocate each specimen into one of three populations (Euro-American, African American, Native American) using the of missing data could be estimated. We began with a list of 14 standard measurements (GOL, FRC, PAC, OCC, XCB, ASB, AUB, ZYB, MAB, deletion for cases with less than nine variables resulted in a sample of 372 individuals with 247 missing data points (5208 total cells or 5% a After Martin (1928). missing data). These data were estimated using the EM algorithm in Systat v. 11.0. Missing data cells for the Seminole and St. Marks ceme- 3 | RESULTS tery individuals were not estimated. The complete data set was imported into Fordisc 3.1. The same analytical procedure was then Analyses using the Contemporary database analyses are presented in used as described above, thus assuring consistency of methodology Tables 3 and 4. Results for the Morton Collection’s Seminole crania across the three data sets. To consider the reliability of allocations (Table 3) indicate a Native American ancestry for all seven male individ- between analysis runs, we used Cohen’s generalized j following Fleiss uals. In general, the step-wise selection process was able to identify (1981); see also Urbanova et al. (2014). discriminant models for all specimens with a suﬃcient number of 6 | GELLER T A B LE 3 AND STOJANOWKSI Comparison of Seminole crania to the “Contemporary” sample Typicality prob. ID Predicted Group F v Rank Post. Prob. D2 % Corr. No Vars. VCVM p 604 Native American .88 .85 .64 .66 8.6 74.4 14 .01 Euro-American .74 .70 .56 .21 10.9 African American .66 .62 .52 .13 11.8 74.4 14 .01 74.4 9 .01 74.4 14 .01 74.4 13 .01 74.1 13 .01 74.4 14 .01 Variables: ASB, AUB, EKB, FRC, GOL, MAB, NLH, NPH, OBB, OBH, OCC, PAC, XCB, ZYB 698 Native American .82 .79 .53 .46 9.6 Euro-American .79 .76 .66 .38 10.0 African-American .67 .63 .52 .16 11.7 Variables: ASB, AUB, EKB, FRC, GOL, MAB, NLH, NPH, OBB, OBH, OCC, PAC, XCB, ZYB 707 Native American .78 .74 .50 .78 10.3 African American .50 .45 .37 .12 14.0 Euro-American .48 .43 .36 .10 14.3 Variables: ASB, EKB, FRC, GOL, MAB, NLH, NPH, OBB, OBH 730 Native American .87 .89 .66 .87 8.2 African American .61 .56 .46 .10 12.5 Euro-American .42 .36 .34 .03 15.2 Variables: ASB, AUB, EKB, FRC, GOL, MAB, NLH, NPH, OBB, OBH, PCC, PAC, XCB, ZYB 732 Native American .84 .84 .62 .66 8.4 African American .71 .67 .54 .25 10.3 Euro-American .55 .51 .40 .09 12.3 Variables: ASB, AUB, EKB, FRC, GOL, MAB, NLH, NPH, OBB, OBH, OCC, PAC, XCB 754 Native American .55 .51 .33 .80 12.2 African American .30 .26 .26 .13 15.9 Euro-American .23 .19 .21 .07 17.3 Variables: ASB, AUB, EKB, FRC, GOL, MAB, NLH, OBB, OBH, OCC, PAC, XCB, ZYB 1105 Native American .99 .99 .91 .79 4.8 African American .92 .91 .87 .19 7.6 Euro-American .68 .64 .45 .03 11.6 Variables: ASB, AUB, EKB, FRC, GOL, MAB, NLH, NPH, OBB, OBH, OCC, PAC, XCB, ZYB % Corr. 5 the overall classiﬁcation success rate for the cross-validation, leave-one-out procedure. No Vars. 5 the number of variables selected, if variable list is alphabetized then no stepwise procedure was needed, otherwise order as determined by stepwise method in FORDISC that maximized the cross-validation rate. VCVM p value 5 the homogeneity of covariance matrix test of Kullback. Bold entries indicate posterior probabilities greater than .66. variables (13 or 14), high typicality probabilities, and moderate to high signiﬁcant; however, none were less than 1 3 1026 and the per- classiﬁcation success rates. However, all Kullback tests were signiﬁcant formance of the allocation may not be aﬀected (Ousley & Jantz, 2012, indicating a violation of the assumption of equality of covariance matri- pp. 317). ces. This does not necessarily invalidate the allocations, however (see Ousley & Jantz, 2005, pp. 54; Ousley & Jantz, 2012, pp. 317). Analyses using the Howells database are presented in Tables 5 and 6. Six of seven Seminole crania allocate as Native American (Table Results for the St. Marks cemetery burials (Table 4) were more 5). The typicality and posterior probabilities indicate a strong model ﬁt mixed. Six of 10 individuals allocated as Euro-American, three as Afri- with very high cross validation allocation success rates for most cases. can American, and one as Native American. Posterior probabilities In fact, when the posterior probabilities are combined for the Arikara were low for several cases that likely reﬂect the higher rate of cross and Santa Cruz samples (see bold entries in Table 5) there is very validation error between African American and Euro-American individ- strong support for the Native American identity of these individuals. uals, which approached 30 to 40%. This rate of error indicates limited The single exception is the individual numbered 730,2 who allocated as separation between African and Euro-American samples, and could reﬂect hybridity among these populations or errors of classiﬁcation in the Angel archives data. The Native American sample, however, was well diﬀerentiated from both African American and Euro-American samples in the graphical Fordisc output. Another factor to consider is that the number of variables recorded was low (n 5 5) for several specimens, reﬂecting poor preservation. As with the analyses of the Seminole crania most of the Kullback tests of variance homogeneity were 2 Morton originally labeled crania with a number. After his death in 1851, the collection was purchased by the Academy of Natural Sciences of Philadelphia (ANSP). In 1966, ANSP loaned the collection to the University of Pennsylvania’s Museum of Archaeology and Anthropology. Penn Museum amended Morton’s cataloguing system by adding the preﬁx L-606; the “L” designated that the collection was on loan. When the loan was formally gifted to Penn Museum in 1997, the catalogue number changed again; the L became 97. For example, 730, today is catalogued as 97-606-730. GELLER AND T A B LE 4 | STOJANOWKSI Comparison of St. Marks crania to the “Contemporary” sample Typicality prob. ID 3 7 Predicted Group F v Rank Post. Prob. D2 % Corr. No Vars. VCVM p 72.4 10 .01 74.3 9 .01 71.8 5 .42 74.4 12 .01 71.8 5 .42 72.4 10 .02 72.4 10 .01 73.5 11 .01 71.8 5 .22 72.9 11 .01 African American .09 .07 .09 .53 17.2 Euro-American .08 .07 .09 .46 17.5 Native American .01 .00 .01 .01 25.7 Variables: AUB, EKB, FRC, GOL, NLH, OBH, OCC, PAC, XCB, ZYB 6 Euro-American .86 .85 .74 .61 4.8 African American .78 .76 .76 .38 5.8 Native American .12 .10 .12 .00 14.8 Variables: AUB, EKB, GOL, MAB, OBH, OCC, PAC, XCB, ZYB 9 Euro-American .97 .97 .96 .55 0.9 African American .93 .93 .95 .42 1.4 Native American .27 .25 .25 .03 6.6 Variables: AUB, GOL, OCC, PAC, XCB 10 African American .97 .97 .94 .72 4.7 Euro-American .90 .88 .77 .28 6.6 Native American .31 .27 .19 .01 14.5 Variables: AUB, EKB, FRC, GOL, MAB, NLH, OBB, OBH, OCC, PAC, XCB, ZYB 13 Native American .57 .56 .46 .79 3.9 African American .24 .23 .26 .18 6.9 Euro-American .11 .10 .13 .06 Variables: AUB, GOL, OCC, PAC, XCB 14 Euro-American .85 .83 .76 .86 5.8 African American .52 .49 .54 .14 9.4 Native American .10 .08 .12 .00 16.7 Variables: AUB, EKB, FRC, GOL, NLH, OBH, PAC, XCB, ZYB 15 Euro-American .90 .89 .80 .72 5.0 African American .75 .73 .79 .27 6.9 Native American .24 .21 .15 .01 13.2 Variables: AUB, EKB, FRC, GOL, NLH, OBH, OCC, PAC, XCB, ZYB 16 Euro-American .72 .70 .65 .50 8.2 African American .67 .64 .66 .38 8.8 Native American .47 .44 .33 .12 11.1 Variables: AUB, EKB, FRC, GOL, MAB, NLH, OBH, OCC, PAC, XCB, ZYB 17 Euro-American .15 .14 .15 .74 8.4 African American .07 .06 .10 .26 10.5 Native American .00 .00 .01 .00 20.7 Variables: AUB, GOL, OCC, PAC, XCB 18 African American .81 .80 .73 .58 7.1 Euro-American .76 .73 .65 .40 7.8 Native American .31 .27 .19 .03 13.4 Variables: AUB, EKB, FRC, GOL, MAB, NLH, OBB, OBH, OCC, PAC, XCB % Corr. 5 the overall classiﬁcation success rate for the cross-validation, leave-one-out procedure. No Vars. 5 the number of variables selected, if variable list is alphabetized then no stepwise procedure was needed, otherwise order as determined by stepwise method in FORDISC that maximized the cross-validation rate. VCVM p 5 the homogeneity of covariance matrix test of Kullback. Bold entries indicate posterior probabilities greater than .66. African American, overwhelmingly so based on the posterior probabil- Typicality and posterior probabilities were generally lower than those ity. Given how much of an outlier this individual is he may represent a for the Seminole cases, despite using the same comparative database. “Black Seminole,” an interpretation supported by macromorphoscopic Cross validation success rates were also more variable and slightly variation (after Hefner, 2009). None of the Kullback tests indicated sig- lower than for the Seminole comparisons. None of the Kullback tests niﬁcant diﬀerences in the sample covariance matrices. indicated signiﬁcant diﬀerences in the sample covariance matrices. Results for the St. Marks cemetery cases using the Howells Analyses using the Forensic Data Bank are presented in Tables 7 database (Table 6) indicate that seven of 10 individuals allocated as and 8. Allocation of the Seminole crania (Table 7) becomes more com- Euro-American, two as African American, and one as Native American. plicated inasmuch as three individuals were identiﬁed as African 8 | GELLER T A B LE 5 AND STOJANOWKSI Comparison of Seminole crania to the Howells data set Typicality prob. ID Predicted Group F v Rank Post. Prob. D2 % Corr. No Vars. VCVM p 604 Santa Cruz .96 .96 .83 .88 4.4 87.5% 11 .97 Arikara .10 .70 .65 .10 8.7 87.5% 11 .99 87.5% 11 .99 87.5% 11 .99 87.5% 11 .99 84.9% 7 .74 87.9% 9 .78 Black .45 .40 .53 .02 11.6 White .07 .05 .08 .00 20.1 Variables: PAC, MAB, AUB, DKB, NLB, BNL, ASB, BPL, EKB, GOL, JUB 698 Arikara .81 .78 .54 .83 7.3 Santa Cruz .55 .50 .23 .17 10.4 19th c. White .03 .01 .02 .00 23.6 19th c. Black .01 .01 .08 .00 25.8 Variables: PAC, MAB, AUB, NLB, ASB, BNL, BPL, EKB, GOL, JUB 707 Arikara .68 .63 .30 .85 8.9 Santa Cruz .36 .31 .10 .12 12.8 19th c. White .18 .14 .23 .03 16.0 19th c. Black .08 .06 .19 .01 19.2 Variables: PAC, MAB, AUB, DKB, NLB, ASB, BNL, BPL, EKB, GOL, JUB 730 19th c. Black .24 .19 .36 .74 14.8 Arikara .12 .08 .03 .16 17.9 Santa Cruz .10 .06 .02 .10 18.8 19th c. White .02 .01 .02 .00 25.4 Variables: PAC, MAB, AUB, DKB, NLB, ASB, BNL, BPL, EKB, GOL, JUB 732 Arikara .73 .68 .42 .49 8.3 Santa Cruz .72 .68 .39 .47 8.4 19th c. Black .25 .20 .36 .02 14.7 19th c. White .23 .19 .28 .02 14.9 Variables: PAC, MAB, AUB, DKB, NLB, ASB, BNL, BPL, EKB, GOL 754 Santa Cruz .21 .18 .04 .90 10.1 Arikara .05 .04 .02 .07 15.1 19th c. Black .02 .02 .09 .02 17.3 19th c. White .00 .00 .01 .00 26.3 5.0 Variables: PAC, MAB, NLH, AUB, ASB, EKB, OBH 1105 Arikara .86 .84 .70 .65 Santa Cruz .74 .70 .48 .33 6.4 19th c. White .21 .18 .29 .01 12.6 19th c. Black .20 .16 .37 .01 13.0 Variables: PAC, BPL, BNL, NLB, ZYB, DKB, XCB, FRC, NPH % Corr. 5 the overall classiﬁcation success rate for the cross-validation, leave-one-out procedure. No Vars. 5 the number of variables selected, if variable list is alphabetized then no stepwise procedure was needed, otherwise order as determined by stepwise method in FORDISC that maximized the cross-validation rate. VCVM p 5 the homogeneity of covariance matrix test of Kullback. Bold entries indicate posterior probabilities greater than .66. American despite continually strong cross validation success rates. the Kullback tests indicated signiﬁcant diﬀerences in the sample covari- None of the Kullback tests indicated signiﬁcant diﬀerences in the sam- ance matrices. ple covariance matrices. Comparison of the typicality and posterior probabilities in Tables 5 and 7 suggests a stronger ﬁt with the Howells 4 | DISCUSSION data set samples than with the Forensic Data Bank. Of particular interest is the individual labeled 730 who allocated to an African American This article uses comparative craniometric analyses to consider how sample in both comparisons. 19th century Seminole crania from the Morton Collection allocate with Results for the St. Marks cemetery (Table 8) indicate that eight of respect to biological diversity in 18th, 19th, and 20th century America. ten individuals allocated as Euro-American, one as African American, As a point of comparison we also included individuals from a 19th cen- and one as Native American. Most individuals demonstrated an tury military cemetery from Wakulla County, Florida. Results from the increase in posterior and typicality probabilities in the Forensic Data diﬀerent analysis runs are summarized in Table 9. In general, the cra- Bank comparisons. Cross-validation success rates were mixed, how- niometric analyses performed as expected. The majority of Seminole ever, and generally lower than those seen in Tables 4 and 6. None of crania allocated as Native American and the majority of St. Marks GELLER AND T A B LE 6 | STOJANOWKSI Comparison of St. Marks crania to the Howells data set Typicality prob. ID Predicted Group 3 19th c. White .33 .28 .41 19th c. Black .19 .14 .32 F v Rank Post. Prob. D2 % Corr. No Vars. VCVM p .79 13.2 85.7% 11 .88 .20 15.9 85.7% 13 .99 70.7% 7 .77 87.5% 17 .99 70.7% 7 .77 77.0% 9 .62 85.7% 11 .88 88.7% 11 .80 71.8% 5 .18 88.7% 12 .97 Arikara .05 .03 .02 .01 21.7 Santa Cruz .01 .00 .02 .00 28.1 Variables: PAC, NLB, AUB, BPL, BNL, EKB, GOL, OBH, NLH, XCB 6 19th c. Black .42 .35 .52 .60 14.4 19th c. White .33 .27 .46 .32 15.6 Santa Cruz .19 .14 .06 .07 18.6 Arikara .04 .02 .02 .00 25.3 Variables: AUB, BBH, BPL, EKB, FOL, GOL, MAB, NLB, OBH, OCC, PAC, XCB, ZYB 9 19th c. White .86 .85 .87 .77 3.4 19th c. Black .50 .47 .53 .15 6.6 Arikara .32 .28 .23 .06 8.6 Santa Cruz .20 .17 .08 .02 10.3 Variables: AUB, BBH, FOL, GOL, OCC, PAC, XCB 10 19th c. White .62 .54 .71 .84 15.8 Santa Cruz .33 .23 .06 .07 20.8 Arikara .32 .23 .09 .06 21.0 19th c. Black .25 .17 .45 .03 22.3 Variables: AUB, BBH, BPL, DKB, EKB, FOL, FRC, GOL, MAB, NLB, NLH, OBB, OBH, OCC, PAC, XCB, ZYB 13 Arikara .46 .43 .30 .70 7.0 Santa Cruz .29 .26 .15 .28 8.9 19th c. White .06 .05 .10 .02 14.1 19th c. Black .03 .02 .11 .01 16.7 Variables: AUB, BBH, FOL, GOL, OCC, PAC, XCB 14 19th c. White .15 .12 .17 .79 14.1 Santa Cruz .06 .04 .17 .16 17.3 19th c. Black .03 .02 .12 .06 19.4 Arikara .01 .00 .02 .00 24.7 Variables: PAC, NLH, EKB, AUB, ZYB, FOL, OBH, XCB 15 19th c. White .12 .09 .13 .99 17.8 Santa Cruz .00 .00 .02 .00 31.1 Arikara .00 .00 .02 .00 31.3 19th c. Black .00 .00 .01 .00 33.0 Variables: PAC, NLB, AUB, BPL, BNL, EKB, GOL, OBH, NLH, XCB, ZYB 16 19th c. White .23 .19 .35 .64 14.9 Santa Cruz .13 .09 .02 .17 17.6 Arikara .13 .09 .05 .15 17.8 19th c. Black .06 .04 .19 .04 20.2 Variables: PAC, MAB, AUB, DKB, NLB, BNL, GOL, XCB, NLH, OBH 17 19th c. White .06 .05 .10 .50 10.9 19th c. Black .06 .05 .12 .42 11.3 Santa Cruz .01 .01 .02 .05 15.6 Arikara .01 .01 .02 .03 16.6 Variables: PAC, AUB, XCB, GOL, BBH 18 19th c. Black .82 .79 .89 .61 8.0 19th c. White .74 .70 .79 .37 9.0 Santa Cruz .28 .22 .08 .02 15.4 Arikara .22 .16 .07 .01 16.7 Variables: PAC, MAB, AUB, NLB, EKB, OBB, BNL, BPL, GOL, OBH, NLH, XCB % Corr. 5 the overall classiﬁcation success rate for the cross-validation, leave-one-out procedure. No Vars. 5 the number of variables selected, if variable list is alphabetized then no stepwise procedure was needed, otherwise order as determined by stepwise method in FORDISC that maximized the cross-validation rate. VCVM p 5 the homogeneity of covariance matrix test of Kullback. Bold entries indicate posterior probabilities greater than .66. 9 10 | T A B LE 7 GELLER Predicted Group 604 STOJANOWKSI Comparison of Seminole crania to the Forensic Data Bank Typicality prob. ID AND F v Post. Prob. Rank D2 % Corr. No Vars. VCVM p 88.6% 13 .98 88.6% 13 .98 89.0% 13 .99 90.8% 14 .99 85.7% 12 .98 84.3% 12 .99 87.7% 11 .99 Black Male .86 .85 .82 .63 8.0 Amerind Male .78 .75 .69 .33 9.3 White Male .42 .38 .39 .04 13.8 Variables: NLB, AUB, BPL, BNL, OBH, DKB, EKB, ZYB, XCB, FOB, FRC, MAB, WFB 698 Amerind Male .56 .56 .45 .48 13.0 White Male .39 .36 .33 .30 14.3 Black Male .30 .26 .35 .14 15.8 Variables: NLB, AUB, BPL, BNL, OBH, DKB, EKB, ZYB, XCB, FOB, FRC, MAB, WFB 707 Amerind Male .38 .33 .37 .70 14.7 White Male .21 .18 .19 .17 17.5 Black Male .19 .15 .22 .13 18.2 Variables: NLB, AUB, BPL, BNL, OBH, DKB, EKB, ZYB, XCB, FOB, FRC, MAB, MDH 730 Black Male .36 .36 .29 .85 16.5 Amerind Male .19 .13 .25 .15 20.0 White Male .01 .00 .01 .00 33.5 Variables: ZYB, BBH, EKB, NLB, ZMB, ASB, AUB, MDH, MAB, BNL, BPL, OBH, DKB, GOL 732 White Male .90 .90 .90 .67 6.4 Black Male .73 .71 .70 .19 8.9 Amerind Male .70 .66 .64 .14 9.5 Variables: NLB, BBH, BPL, BNL, AUB, OBH, DKB, EKB, MAB, FOB, FRC, XCB 754 Black Male .21 .18 .28 .98 16.3 Amerind Male .03 .02 .15 .02 23.7 White Male .01 .00 .01 .00 29.4 Variables: MAB, AUB, BBH, DKB, OBH, BNL, FOB, ZTB, FRC, XCB, OBB, EKB 1105 Amerind Male .90 .88 .89 .43 5.9 Black Male .89 .88 .85 .43 5.9 White Male .72 .70 .65 .14 8.1 Variables: NLB, AUB, BPL, BNL, OBH, DKB, EKB, MAB, FRC, XCB % Corr. 5 the overall classiﬁcation success rate for the cross-validation, leave-one-out procedure. No Vars. 5 the number of variables selected, if variable list is alphabetized then no stepwise procedure was needed, otherwise order as determined by stepwise method in FORDISC that maximized the cross-validation rate. VCVM p 5 the homogeneity of covariance matrix test of Kullback. Bold entries indicate posterior probabilities greater than .66. cemetery soldiers allocated as Euro-American regardless of the com- allocate as Native American and seven of ten St. Marks soldiers allo- parative database used. Deviations from these expectations are dis- cate as Euro-American. cussed further below and provide some of the more compelling results from this study. The Contemporary database was constructed from a variety of sources to mitigate a typological approach and focus on the speciﬁc It is important to note that the comparative samples represent a populations from which the Morton Collection Seminole and St. Marks loose temporal sequence. The Contemporary sample includes individu- cemetery burials are thought to derive given archival documentation. als roughly of the same antiquity or slightly older than the Morton Col- For example, the Contemporary database uses late pre-contact and lection and St. Marks cemetery crania; the Howells database includes early historic period data from Native American sites in Georgia to sup- individuals primarily from the 19th century; and the Forensic Data plement the two North American series provided by Howells. The geo- Bank includes individuals primarily from the 20th century. None of the graphic ﬁt is strong but the samples are older than the 19th century samples is a perfect replication of the target population. The Howells Seminole crania included here. Importantly, the Georgia data pre-date data set is the standard used in many analyses of craniometric varia- the complex processes of Seminole ethnogenesis, which involved tion. For this study, it maximizes temporal ﬁt between target and demographic collapse, population aggregations, and long-distance relo- source populations and thereby reduces the eﬀects of secular trends. cations during the 18th century (Sattler, 1996; Stojanowski 2010; Stur- Use of 19th century White and Black samples in Fordisc is appropriate tevant & Cattelino, 2004; Wickman, 1999). Comparative samples of here; however, the geographic ﬁt is poor for the Native American sam- African and Euro-Americans derive from a diverse array of contexts ples (California Chumash, South Dakota Arikara) provided by Howells. from the eastern U.S., largely representing colonial period sites from Despite this, the Howells allocations are robust. Typicality and poste- the 18th and 19th centuries. For the Seminole allocations, all seven rior probabilities are high. Six of seven Morton Collection Seminole individuals allocated as Native American with strong typicality and GELLER AND T A B LE 8 | STOJANOWKSI Comparison of St. Marks crania to the Forensic Data Bank Typicality prob. ID 3 11 Predicted Group F v Rank Post. Prob. D2 % Corr. No Vars. VCVM p 88.0% 10 .98 79.6% 7 .37 69.9% 9 .82 88.7% 12 .99 69.6% 10 .36 78.4% 9 .86 88.0% 10 .98 86.2% 13 .99 68.0% 8 .80 86.0% 12 .99 White Male .57 .55 .57 .73 8.8 Black Male .37 .35 .36 .23 11.1 Amerind Male .16 .14 .17 .04 14.9 Variables: BPL, BNL, NLB, AUB, OBH, ZYB, EKB, FOB, XCB, FRC 6 White Male .32 .31 .31 .52 8.2 Black Male .27 .25 .27 .36 9.0 Amerind .15 .13 .28 .12 11.1 Variables: XCB, NLB, BBH, FOB, WFB, OBH, ZYB 9 White Male .97 .97 .95 .89 2.9 Black Male .62 .61 .55 .10 7.3 Amerind Male .27 .24 .33 .01 11.5 Variables: AUB, BBH, FOB, FOL, GOL, OCC, PAC, WFB, XCB 10 White Male .50 .48 .46 .99 11.6 Black Male .01 .00 .01 .00 29.4 Amerind Male .01 .00 .04 .00 29.4 Variables: NLB, AUB, BPL, BNL, OBH, DKB, ZYB, EKB, XCB, FOB, FRC, MAB 13 Amerind Male .41 .31 .67 .61 11.6 White Male .25 .24 .23 .33 12.8 Black Male .10 .09 .14 .06 16.4 Variables: AUB, BBH, FOB, FOL, GOL, OCC, UFBR, WFB, XCB 14 White Male .31 .30 .29 .97 10.7 Black Male .05 .04 .07 .03 17.6 Amerind Male .01 .01 .04 .00 23.8 Variables: FOB, XCB, EKB, AUB, ZYB, BBH, OBH, BNL, FRC 15 White Male .58 .56 .57 .99 8.7 Amerind Male .02 .02 .08 .00 21.8 Black Male .00 .00 .01 .00 27.9 Variables: BPL, BNL, NLB, AUB, OBH, ZYB, EKB, FOB, XCB, FRC 16 White Male .35 .32 .32 .67 14.9 Black Male .25 .21 .24 .26 16.8 Amerind Male .15 .12 .26 .08 19.2 Variables: NLB, XCB, FOB, BBH, MAB, OBH, AUB, DKB, BNL, FRC, ZYB, OBB, NLH 17 White Male .39 .38 .38 .63 8.6 Black Male .31 .29 .34 .37 9.6 Amerind Male .05 .03 .08 .01 17.5 Variables: AUB, BBH, GOL, OCC, UFBR, PAC, WFB, XCB 18 Black Male .10 .08 .13 .62 19.4 White Male .07 .05 .05 .29 20.9 Amerind Male .04 .03 .12 .09 23.3 Variables: NLB, BBH, BPL, BNL, AUB, OBH, EKB, OBB, FOB, FRC, MAB, XCB % Corr. 5 the overall classiﬁcation success rate for the cross-validation, leave-one-out procedure. No Vars. 5 the number of variables selected, if variable list is alphabetized then no stepwise procedure was needed, otherwise order as determined by stepwise method in FORDISC that maximized the cross-validation rate. VCVM p 5 the homogeneity of covariance matrix test of Kullback. Bold entries indicate posterior probabilities greater than .66. posterior probabilities. The allocations for the St. Marks soldiers were African American (burial 18), regardless of the database to which each less robust; six of ten were identiﬁed as Euro-American. was compared (see below). The largely 20th century Forensic Data Bank samples suﬀer from With these results in mind, the rest of the discussion focuses on both a poor geographic and temporal ﬁt for the Morton Collection two key aspects of the analyses. First, we discuss the overall patterning Seminole, and as shown in Table 9 produce the most divergent results of the craniometric analysis and how the results change for each sam- with a greater percentage of Morton Collection crania allocating as ple with respect to the populations to which they are compared. Sec- African American. Results for the St. Marks burials were more consist- ond, we discuss those cases that deviate from the expected ancestral ent with eight of ten individuals allocating as Euro-American. Two population aﬃnities and contrast methodological and biocultural exceptions consistently allocated as Native American (burial 13) and aspects of these results. 12 | GELLER Summary of allocation results for the Seminole and St. Marks crania T A B LE 9 Morton seminole Contemporary Howells FDB 604 NA (.88, .66) NA (.96, .88) AA (.86, .63) AND STOJANOWKSI allocate as African American as one moves closer to the present, while individuals identiﬁed as Euro-American increasing allocate as such as one moves closer to the present. Previous work on comparative craniometry might interpret low repeatability as a methodological issue—that the dependence of the 698 NA (.82, .46) NA (.81, .83) NA (.56, .48) 707 NA (.78, .78) NA (.68, .85) NA (.38, .70) acy of comparative craniometric analysis as a whole. And while surely 730 NA (.87, .87) AA (.24, .74) AA (.36, .85) some errors have crept into this analysis, three outlier cases are of par- 732 NA (.84, .66) NA (.73, .49) EA (.90, .67) 754 NA (.55, .80) NA (.21, .90) AA (.21, .98) 1105 NA (.99, .79) NA (.86, .65) NA (.90, .43) St. Marks Cemetery allocation on the speciﬁc samples used reﬂects instability and inaccur- ticular interest: burials 13 and 18 from the St. Marks cemetery and individual 730 from the Morton Collection. With respect to the St. Marks soldiers, burial 13 was identiﬁed as Native American and burial 18 was identiﬁed as African American in all three allocation runs. There is little indication of individuals of a non-White ancestry in the published descriptions of these crania or in the archival roles from the St. 3 AA (.09, .53) EA (.33, .79) EA (.57, .73) 6 EA (.86, .61) AA (.42, .60) EA (.32, .52) 9 EA (.97, .55) EA (.86, .77) EA (.97, .89) 10 AA (.97, .72) EA (.62, .84) EA (.50, .99) 13 NA (.57, .79) NA (.46, .70) NA (.41, .61) 14 EA (.85, .86) EA (.15, .79) EA (.31, .97) 15 EA (.90, .72) EA (.12, .99) EA (.58, .99) 16 EA (.72, .50) EA (.23, .64) EA (.35, .67) of Andrew Jackson’s invading force (Dailey et al., 1972). Although the 17 EA (.15, .74) EA (.06, .50) EA (.39, .63) allocations of burials 13 and 18 may be in error, several things point to 18 AA (.81, .58) AA (.82, .61) AA (.10, .62) their validity. First, the allocations are consistent across databases with Marks cemetery (Dailey et al., 1972). Although burial 13 was too poorly preserved for assessment, macromorphoscopic variation supports the Note. In Howells column NA indicates either an allocation to the Arikara or to the Santa Cruz samples. Bold entries indicate an allocation that did change not across analysis runs. The parentheses are (F typicality probability, posterior probability). NA 5 Native American, EA 5 Euro-American, AA 5 African American. African American identity of burial 18 (summed score 5 3; after Hefner (2009)). That an African American or Native American may have been buried at Fort St. Marks may not be so unusual since Florida was a contested frontier zone during the 19th century. Large, enslaved African populations were recorded and Creek warriors comprised the majority high probability values and variable numbers that are comparable to those for the other individuals. There is no reason to discount these results on methodological grounds. Second, both men were buried with plain, handmade bone buttons, a contrast to the majority of burials containing pewter buttons with military designs (see Dailey et al., 1972, pp. 4). Although inconclusive, this serves as independent evi- With respect to the general patterning of allocations presented in dence that burials 13 and 18 may have had a status distinct from other Table 9, it is interesting to note that only three of seven Seminole cra- individuals interred in the cemetery. Third, military posts in Florida dur- nia produce the same population allocation regardless of the database ing this time period were multiethnic. For example, at the mid-19th used. Instead, Seminole individuals allocate as African Americans as the century Fort Brooke cemetery (also in Florida), it was observed that, “a comparative samples become more modern in age (0 individuals in the standard procedure for corpse disposal, in which the dead were buried Contemporary sample, one in the Howells data set, and four in the For- in coﬃns in a cemetery, was followed regardless of the racial origin of ensic Data Bank). To the contrary, the St. Marks cemetery individuals— the deceased” (Piper, Hardin, & Piper, 1982, pp. 130, emphasis added). comprised of mostly White (Euro-American) soldiers—show relative As such, it is entirely possible the craniometric analyses are accurate in consistency in the results. Seven of ten individuals do not change pop- their assessment of population aﬃnity. However, given the lack of sup- ulation allocation across the three databases and there is only a single porting documentation little more can be inferred from these analyses. diﬀerence between the Howells and Forensic Data Bank results. In The same cannot be said for individual 730 from the Morton contrast, only three of seven Seminoles return the same allocation Collection, an outlier identiﬁed as African American in both the depending on whether the Howells or FDB database is used. The gen- Howells and FDB analyses. As with burial 18 from St. Marks, individual eralized j for the Morton Collection crania measured across all three 730 preserves cranial macromorphoscopic variation consistent with an databases was zero (j 5 .000, p 5 .999), indicating very poor repeat- African American identity (summed score 5 3; after Hefner (2009)). He ability. Most of this results from the inclusion of the Forensic Data was a male who died between the ages of 25 and 35. Although he dis- Bank. Repeatability for the St. Marks crania was moderate across the plays no ante- or perimortem trauma, this individual was killed on 25 three databases and signiﬁcantly diﬀerent from 0 (j 5.565, p 5 .019). December 1837 at the Battle of Lake Okeechobee. A month later, U.S. Repeatability was lowest for the Contemporary-Howells database Army Surgeon Eugene H. Abadie collected his cranium and a second comparisons, the opposite of that observed for the Seminole crania. In cranium for Morton. “I collected two ﬁne skulls which were the only other words, individuals labeled as Seminole by Morton increasingly two out of the 12 Indians killed which could be taken the others being GELLER AND | STOJANOWKSI 13 very oﬀensive,” Abadie (1838) explained in a letter. When examined by Even prior to the Battle of Lake Okeechobee, the relationship PLG, soft tissue still adhered to the palate of both of these individuals; between Seminole Indians and their Black allies had troubled members skeletonization was largely though not entirely complete when Abadie of the U.S. government and military. Major General Thomas Jesup, encountered the bodies, but exposure after death hastened the pro- Commander of the U.S. Army during the Second Seminole War, vocal- cess. Abadie also identiﬁed these individuals as warriors. After receiv- ized this concern in an 1836 letter to Acting Secretary of War B.F. ing the crania, Morton inked “Seminole” across both of their foreheads. Butler: “This, you may be assured, is a negro, not an Indian war; and if Based on this we propose that individual 730 was a Black Seminole it be not speedily put down, the South will feel the eﬀects of it on their and his presence in the Morton Collection provides important insights slave population before the end of the next season” (as quoted by into the production of race as a category of social identity during the Montgomery, 1839, pp. 269). But, Jesup also realized that Seminole th 19 century, a process known as racialization. Indians would not relocate without their Black allies. So he went on to Based on similarities between naming practices, Rebecca Bateman broker various treaties, starting in March 1837 with the Camp Dade (2002) has argued that people from the Kongo-Angola region in Africa accord, in support of westward immigration of Seminole Indians and were well represented amongst Blacks living in 19th century Florida. Blacks (Mulroy, 2007, pp. 46-50). But, there was still much ambiguity Their paths often crossed with Seminole Indians, whose own ethno- built into treaties that functioned to locate Native and African Ameri- genesis occurred after arriving in northern Florida at the start of the cans in the developing nation, both physically and conceptually. 18th century (Sturtevant & Cattelino, 2004, pp. 429). Prior to 1822 Namely questions remained about the status of Blacks removed to when Florida oﬃcially became a U.S. territory, this frontier held the Indian Territory as “bona ﬁde property” of Seminole Indians or freed- promise of emancipation for enslaved Blacks, as historians of the colo- men. Whether strategic or unconscious, the nature and complexity of nial period have discussed extensively (e.g., Aptheker, 1939; Corbett, Seminole Indians and Black Seminoles’ interactions went undiscussed 1974; Landers, 1990; Missall & Missall, 2004; Riordan, 1996; Schafer, in oﬃcial discourse. Rather, during this period of nation-building, one’s 1993; Twyman, 1999). Its sheltering swamps acted as refuge and pro- access to or denial of American citizenship was predicated on observ- vided the backdrop for Black Seminole ethnogenesis.3 Terrence Weik’s able, measurable, and clear-cut boundaries between racial groups. In (1997, 2007, 2009) investigations of colonial Florida have documented this regard, Morton’s craniometric studies, which reiﬁed race as the archaeological evidence for these experiences and processes. But, bounded, natural, and hierarchical, were particularly inﬂuential. as he also recognizes, sociopolitical relations between Blacks and Semi- For those peoples such as Black Seminoles whose hybridity deﬁed nole Indians were complex and multi-faceted—contoured by econom- racial categorizations, public policies and scientiﬁc study worked to ics, procreation, desire, and combat (see also Katz, 1986; Littleﬁeld, erase them from the national consciousness and oﬃcial histories. For 1977; Mock, 2012; Mulroy, 2007; Porter, 1996). Some individuals the former, social and legal policies like the one-drop rule for Blacks were brought into the fold of natives’ bands through marriage or adop- and blood quantum for Native Americans were eﬀective. In the case of tion. Others formed independent communities and acted as tributary the latter, Morton’s own scientiﬁc work on hybridity was demonstra- allies. Many Blacks were re-enslaved by the Seminole themselves. tive. Amongst his collection were “Mixed Races,” including two individ- Regardless of the nature of their relationships with Seminole Indi- uals described as “Negro and Indian” (Morton, 1849, pp. vi). ans, Blacks ran the risk of recapture by White Euro-Americans and “Dr. Wilson, who dissected this man, considered him a full-blooded Creeks whose variant of bondage was described as far harsher than Choctaw, but the skull strongly indicates a mixture of the Negro,” he Seminole enslavement (Missall & Missall, 2004; Simmons, 1822). In an wrote of one (Morton, 1849, pp. 76). Morton designated this individual attempt to resist encroachment by White frontier settlers, as well as “Choctaw and Negro?” The question mark signals his categorical confu- the U.S. government’s relocation policies and violent incursions, Semi- sion. Such an identiﬁcation also contrasts with the individual he labeled nole Indians and Blacks fostered strong military alliances. Warriors 730 and Seminole. In this instance, morphological analysis revealed fought and died together at all battles of the Second Seminole War, as nothing of the decedent’s “Blackness” to Morton. It is not clear many chroniclers of the time remarked. U.S. Army Surgeon Jacob Rhett whether the identiﬁcation was a result of willful ignorance or ﬂawed Motte, for instance, oﬀered a ﬁrsthand account of the Battle of Lake analysis. Okeechobee, which supplied two of the crania analyzed here (see Morton did not begin to explore the issue of hybridity in depth Table 1). “The number of Indians engaged,” he wrote, “were estimated until the end of his life. In the 1847 article “Hybridity in Animals, Con- from two to three hundred warriors and there were with them prob- sidered in Reference to the Question of the Unity of the Human Spe- ably as many negroes” (Motte, 1953, pp. 195). We believe “specimen” cies,” he argued that procreation between individuals of diﬀerent kinds 730, collected during this same battle, is one such individual whose could result in fertile oﬀspring despite their sharing a “natural repug- African heritage was erased by Morton’s call for Indian crania. nance” for each other (Morton, 1847, pp. 210). However, the physician’s interest in hybridity should not be mistaken for acceptance of 3 Here we use “Black Seminole” though Weik prefers African Seminole so as to avoid homogenization and racialization. He (2009, pp. 207) writes, “Although ‘Blackness’ has been historically reclaimed (e.g., ‘Black Power’), it does not explicitly acknowledge the biological and cultural African connections that contributed to African-Native American belief and physiology.” hybrid humans. Rather, Morton (1851) suggested that miscegenation was objectionable because it likely decreased a pure race’s fertility rate. Desire to the contrary was deviant and worthy of juridical punishment and anti-miscegenetic policies, which had existed since the 17th 14 | GELLER AND STOJANOWKSI century but were normalized and nation-wide by the mid-19th century To document the events and materiality of hybrid lives long silenced (Curtin, 2012). While racial categories changed over time, social and is no easy task given the asymmetrical power relations involved in his- legal thought promoted the notion of White superiority and the need tory’s production. It requires a certain amount of strategizing about the to maintain Whites’ racial purity. It is not surprising, then, that certain availability of source material, a sentiment echoed by anthropologist hybrid individuals, like individual 730 included here, may have gone Michel-Rolph Trouillot (1995, pp. 49) who noted, “Silences are inherent undocumented in historic skeletal collections. in history because any single event enters history with some of its consti- In contrast, African and Native American communities neither fos- tuting parts missing.” The biohistoric approach we have laid out here— tered such ideological beliefs nor created regulatory practices about one that brings together comparative craniometric data, archival sources, social and sexual interactions. Our analysis shows that it is diﬃcult to and critical race scholars—can oﬀer an account of the past, in this case diﬀerentiate between African and Native American aﬃliations as the 19th century nation-building, which counters oﬃcial versions. Hence, we comparative samples become more modern in age. This is because do not see our methods as attempting to use the master’s tools to dis- gene ﬂow continued to occur between these two communities in Flor- mantle the master’s house, a strategy that Audre Lorde (1984) sees as ida in ways not observable in Euro-American specimens. That is, disingenuous and ineﬀective. Rather, we seek to reinvent those tools hybridity did not carry as great a social stigma for Blacks and Native entirely. In doing so, we deepen understanding of racialization processes Americans as it did for Whites. During this critical juncture in the 19th that brought races into being and the machinations of racism during a century, Whiteness solidiﬁed and came to undergird a conception of pivotal moment in American history. national citizenship whereas categories of Black and Indian continued Yet, hybridity does not only raise a potential complication to our his- to evolve in response to social mores and the legal realities of minority toric understanding of racialization processes that deﬁned national citi- status. Our results show it was easier to be identiﬁed as White, while zenship. We also see the issue of hybridity as impacting contemporary the categories of non-White blended through time (see also Algee- repatriation practices that rely on racial categorizations. To make amends Hewitt, 2016). Hence, the Black Indian vanished. Recognizing this pro- for the interpersonal violence of colonial expansionism that begot collec- cess remains critical in contemporary medicolegal practice if the sam- tions like the one compiled by Morton and the structural violence result- ples used to deﬁne ranges of biological variation are themselves the ing from its analysis, the Native American Graves Protection and products of multiple centuries of hybridity. Forensic anthropologists are Repatriation Act (NAGPRA) was passed in 1990. Nevertheless, for Native now addressing these issues for Hispanic populations (e.g., Edgar, 2013; American and Hawaiian groups whose ancestors have comprised it, the Hefner, Pilloud, Black, & Anderson, 2015; Hughes, Tise, Trammell, & Collection continues to evoke strong sentiments (see Dumont, 2003; Anderson, 2013; Spradley & Jantz, 2016; Tise, Kimmerle, & Spradley, Pensley, 2005). NAGPRA is silent on the subject of mixed races, how- 2014). Results presented here suggest processes, which are often subtle ever. Institutions receiving federal funds are only obligated to consult and diﬃcult to disentangle, aﬀect many postcolonial sites, cemeteries, with or repatriate to federally-recognized Indian tribes in the U.S. Conse- and samples. quently, contemporary Black Seminole communities, which are currently We see hybridity as no less germane today though for reasons that depart conceptually from Morton. Henry Louis Gates has explored how located in Oklahoma, Texas, Mexico, and the Bahamas (Mock, 2012), have no oﬃcial recourse to repatriate their ancestors. dominant discourses can be appropriated and revised to catalyze socio- In the case of the decedent labeled 730, on 12 October 2015 the political and intellectual change. More speciﬁcally, he has discussed “sig- Penn Museum repatriated his cranium and those of the 15 other Semi- nifyin(g),” a strategy that relies on doubling to trigger sociopolitical noles in the Morton Collection to the Seminole Tribe of Florida. Black transgression or transformation (Gates 1988). Signifyin(g) can involve Seminole communities were not involved in the consultation process. repetition and diﬀerence (i.e., parody) or formal revision. Hybridity is Ethically speaking, their absence may compel native peoples, policy a (1987, pp. one example of signifyin(g). Chicana feminist Gloria Anzaldu makers, and anthropologists to reﬂect on the repatriation process’s 77), for instance, did touch on this double nature of hybridity: “At the shortcomings. Nevertheless, all 16 crania were reinterred at Lake conﬂuence of two or more genetic streams, with chromosomes con- Okeechobee Battleﬁeld, a setting that brings “specimen” 730 full circle stantly ‘crossing over,’ this mixture of races, rather than resulting in an 178 years after a U.S. Army surgeon initially fragmented his head from inferior being, provides hybrid progeny, a mutable, more malleable spe- his body in the name of nation and science. cies.” She continued by stressing that the biological, whether born of corporeal violation or desire, is only one facet of hybridity. Cultural, racial, or ethnic hybridity—she and other feminist writers have ACKNOWLEDGMENTS expounded (e.g., Mohanty, 1988; Visweswaran, 1993)—begs considera- Appreciation goes to Stacey Espenlaub and Janet Monge at Penn tion of existences that are liminal, negotiable, empowering, transgres- Museum for access to archival materials and the Samuel G. Morton sive, situational, and auspicious. We would, however, like to also point Crania Collection. out that while hybridity may be common ground upon which scientists and humanists can tread, the latter are talking about discourse and not R EF ER E N CE S practice or materiality. That is, they attend little to the complexities that Abadie, E. H. (1838). Letter to Samuel G. Morton, 3 February 1838. Samuel George Morton papers. Philadelphia: American Philosophical Society. arise from thinking about biological and cultural interdigitations. GELLER AND STOJANOWKSI Albanese, J., Osley, S. E., & Tuck, A. (2012). Do century-speciﬁc equations provide better estimates of stature? A test of the 19th220th century boundary for the stature estimation feature in Fordisc 3.0. Forensic Science International, 219, 286. Albanese, J., & Sanders, S. R. (2006). Is it possible to escape racial typology in forensic identiﬁcation? In A. Schmitt, E. Cunha, & J. Pinheiro (Eds.), Forensic anthropology and medicine (pp. 281–316). Totowa, NJ: Humana Press. Algee-Hewitt, B. (2016). Population inference from contemporary American craniometrics. American Journal of Physical Anthropology, 160, 604–624. Angel, J. L. (1976). Colonial to modern skeletal change in the U.S.A. American Journal of Physical Anthropology, 45, 723–735. Aptheker, H. (1939). Maroons within the present limits of the United States. Journal of Negro History, 24, 167–184. a, G. (1987). Borderlands/la frontera: The new mestiza. San FranAnzaldu cisco: Aunt Lute Books. Armelagos, G. J., & Van Gerven, D. P. (2003). A century of skeletal biology and paleopathology: Contrasts, contradictions, and conﬂicts. American Anthropologist, 105, 53–64. Barkan, E. (1992). The retreat of scientiﬁc racism: changing concepts of race in Britain and the United States between the World Wars. Cambridge: Cambridge University Press. | 15 tions and the forensic data bank. American Journal of Physical Anthropology, 132, 83–84. Cantwell, A. -M. (2000). “Who knows the power of his bones.” Reburial redux. Annals of the New York Academy of Sciences, 925, 79–119. Cartales, J. (2015). The potential conﬂict between forensic ethnic identiﬁcation and societal interpretation in America. PURE Insights, 4, 1–11. Available at: http://digitalcommons.wou.edu/pure/vol4/iss1/2, accessed 25 September 2016. Corbett, T. G. (1974). Migration to a Spanish imperial frontier in the seventeenth and eighteenth centuries: St. Augustine. Hispanic American Historical Review, 54, 414–430. Cross, P. J., & Wright, R. (2015). The Nikumaroro bones identiﬁcation controversy: First-hand examination versus evaluation by proxy – Amelia Earhart found or still missing? Journal of Archaeological Science Reports, 3, 52–59. Curtin, R. S. (2012). Anti-miscegenation laws. In D. T. Critchlow, & P. R. VanderMeer (Eds.), The Oxford encyclopedia of American political and legal history (pp. 44–46). Oxford, UK: Oxford University Press. Dailey, R. C., Morell, L. R., & Cockrell, W. A. (1972). The St. Marks cemetery (8WA108). Bureau of Historic Sites and Properties, Bulletin No 2. Tallahassee, FL: Department of State, Division of Archives, History & Records Management. Bass, W. M. (2005). Human osteology: a laboratory and ﬁeld manual. Columbia, MO: Missouri Archaeological Society, Inc. de la Cova, C. (2010). Cultural patterns of trauma among 19th-centuryborn males in cadaver collections. American Anthropologist, 112, 589– 606. Bateman, R. B. (2002). Naming patterns in Black Seminole ethnogenesis. Ethnohistory, 49, 227–257. de la Cova, C. (2011). Race, health, and disease in 19th-century-born males. American Journal of Physical Anthropology, 144, 526–537. Beck, L. (1980). Physical anthropology of skeletal remains from Oakland cemetery, Atlanta, Georgia. Unpublished honors thesis. Atlanta, GA: Department of Anthropology, Georgia State University. de la Cova, C. (2012). Trauma patterns in 19th-century-born African American and Euro-American females. International Journal of Paleopathology, 2, 61–68. Belcher, R., Williams, F. L., & Armelagos, G. J. (2002). Misidentiﬁcation of Meroitic Nubians using FORDISC 2.0. American Journal of Physical Anthropology, 34, S42. Berg, G. E., & Ta’ala, S. C. (2015). Biological aﬃnity in forensic identiﬁcation of human skeletal remains. Boca Raton, FL: CRC Press. Birkby, W. H. (1966). An evaluation of race and sex identiﬁcation from cranial measurements. American Journal of Physical Anthropology, 24, 21–27. Blakey, M. (1987). Skull doctors: Intrinsic social and political bias in the history of American physical anthropology. Critical Anthropology, 7, 7–35. Boas, F. (1912). Changes in bodily form of descendants of immigrants. American Anthropologist, 14, 530–562. Dirkmaat, D. C., Cabo, L. L., Ousley, S. D., & Symes, S. A. (2008). New perspectives in forensic anthropology. Yearbook of Physical Anthropology, 51, 33–52. Dumont, C. W. (2003). The politics of scientiﬁc objections to repatriation. Wicazo Sa Review, 18, 109–128. Edgar, H. J. H. (2009). Biohistorical approaches to “race” in the United States: Biological distances among African Americans, European Americans, and their ancestors. American Journal of Physical Anthropology, 139, 58–67. Edgar, H. J. H. (2013). Estimation of ancestry using dental morphological characteristics. Journal of Forensic Sciences, 58, S3–S8. Elliott, M., & Collard, M. (2009). FORDISC and the determination of ancestry from cranial measurements. Biology Letters, 5, 849–852. Brace, C. L. (1995). Region does not mean “race” – reality versus convention in forensic anthropology. Journal of Forensic Sciences, 40, 171–175. Fabian, A. (2010). The skull collectors: race, science, and America’s unburied dead. Chicago: University of Chicago Press. Brace, C. L. (2005). Race” is a four-letter word. New York: Oxford University Press. Fleiss, J. L. (1981). Statistical methods for rates and proportions (2nd edn.). New York: John Wiley and Sons. Brues, A. (1992). Forensic diagnosis of race – general race vs speciﬁc populations. Social Sciences & Medicine, 34, 125–128. Freid, D. L., Spradley, M.,K., Jantz, R. L., & Ousley, S. D. (2005). The truth is out there: How NOT to use FORDISC. American Journal of Physical Anthropology, S40, 103. Buikstra, J. E., & Ubelaker, D. H. (1994). Standards for data collection from human skeletal remains. Fayetteville, AR: Arkansas Archaeological Survey Research. Buxton, L. H. D., Trevor, J. C., & Julien, A. H. (1938). Skeletal remains from the Virgin Islands. Man, 38, 49–52. Gates, H. L. (1988). The signifying monkey: A theory of African American literary criticism. Oxford: Oxford University Press. Byers, S. N. (2010). Introduction to forensic anthropology. New York: Pearson. Geller, P. L. (2015). Hybrid lives, violent deaths: Seminole Indians and the Samuel G. Morton Collection. In Z. Crossland & R. Joyce (Eds.), Disturbing bodies: Perspectives on forensic anthropology (pp. 137–156) Santa Fe, NM: SAR Press. Campbell, A. R., & Armelagos, G. J. (2007). Assessment of FORDISC 3.0’s accuracy in classifying individuals from WW Howell’s popula- Giles, E., & Elliot, O. (1962). Race identiﬁcation from cranial measurements. Journal of Forensic Sciences, 7, 147–157. 16 | Gill, G. W. (1998). Craniofacial criteria in the skeletal attribution of race. In K. J. Reichs (Ed.), Forensic osteology: Advances in the identiﬁcation of human remains (pp. 293–317) Springﬁeld: Charles C. Thomas Publisher, Ltd. Gill, G. W., & Rhine, S. (Ed.). (1990). Skeletal attribution of race. Albuquerque, NM: Maxwell Museum of Anthropology. Gould, S. J. (1978). Morton’s ranking of races by cranial capacity. Science, 200, 503–509. Gould, S. J. (1981). The mismeasure of man. New York: W.W. Norton & Company, Inc. Gravlee, C. C., Bernard, H. R., & Leonard, W. R. (2003a). Heredity, environment, and cranial form: A re-analysis of Boas’s immigrant data. American Anthropologist, 105, 123–138. Gravlee, C. C., Bernard, H. R., & Leonard, W. R. (2003b). Boas’s changes in bodily form: The immigrant study, cranial plasticity, and Boas’s physical anthropology. American Anthropologist, 105, 326–332. Hefner, J. T. (2009). Cranial nonmetric variation and estimating ancestry. Journal of Forensic Sciences, 54, 985–995. Hefner, J. T., Spradley, M. K., & Anderson, B. (2014). Ancestry assessment using random forest modeling. Journal of Forensic Sciences, 59, 583–589. Hefner, J. T., & Ousley, S. D. (2014). Statistical classiﬁcation methods for estimating ancestry using morphoscopic traits. Journal of Forensic Sciences, 59, 883–890. Hefner, J. T., Pilloud, M. A., Black, C. J., & Anderson, B. E. (2015). Morphoscopic trait expression in “Hispanic” populations. Journal of Forensic Sciences, 60, 1135–1139. Herring, D. A., & Swedlund, A. C. (Ed.). (2003). Human biologists in the archives. Cambridge: Cambridge University Press. Horsman, R. (1975). Scientiﬁc racism and the American Indian in the mid-nineteenth century. American Quarterly, 27, 152–168. Howells, W. W. (1973). Cranial variation in man: A study by multivariate analysis of patterns of diﬀerence among recent human populations. Cambridge, MA: Harvard University. Howells, W. W. (1989). Skull shapes and the map: Craniometric analyses in the dispersion of modern homo. Cambridge, MA: Harvard University. Howells, W. W. (1995). Who’s who in skulls: Ethnic identiﬁcation of crania from measurements. Cambridge, MA: Harvard University. Hubbe, M., & Neves, W. A. (2007). On the misclassiﬁcation of human crania: Are there any implications for assumptions about human variation? Current Anthropology, 48, 285. Hughes, C. E., Tise, M. L., Trammell, L. H., & Anderson, B. E. (2013). Cranial morphological variation among contemporary Mexicans: Regional trends, ancestral aﬃnities, and genetic comparisons. American Journal of Physical Anthropology, 151, 506–517. Humphries, A. L. (2011). Craniometric variation in the Caribbean and Latin America as inﬂuenced by the trans-Atlantic slave trade: An examination of the Angolan inﬂuence. MA Thesis. Department of Anthropology, North Carolina State University. Hunt, D. R., & Albanese, J. (2005). History and demographic composition of the Robert J. Terry anatomical collection. American Journal of Physical Anthropology, 127, 406–417. Jantz, R. L. (2001). Cranial change in Americans: 1850-1975. Journal of Forensic Sciences, 46, 784–787. GELLER AND STOJANOWKSI Kakaliouras, A. M. (2008). Leaving few bones unturned: Recent work on repatriation by osteologists. American Anthropologist, 110, 44–52. Kakaliouras, A. M. (2012). An anthropology of repatriation: Contemporary physical anthropological and Native American ontologies of practice. Current Anthropology, 53, S210–S221. Kallenberg, L., & Pilbrow, V. (2012). Using CRANID to test the population aﬃnity of known crania. Journal of Anatomy, 221, 459–464. Katz, W. (1986). Black Indians. New York: Macmillan Children’s Book Group. Keita, S. O. Y. (2007). On Meroitic Nubian crania, Fordisc 2.0, and human biological history. Current Anthropology, 48, 425–427. Klales, A. R., & Kenyhercz, M. W. (2015). Morphological assessment of ancestry using cranial macromorphoscopics. Journal of Forensic Sciences, 60, 13–20. Konigsberg, L. W., Algee-Hewitt, B. F., & Steadman, D. W. (2009). Estimation and evidence in forensic anthropology: Sex and race. American Journal of Physical Anthropology, 139, 77–90. Kosiba, S. (2000). Assessing the eﬃcacy and pragmatism of “race” designation in human skeletal identiﬁcation: A test of Fordisc 2.0. American Journal of Physical Anthropology, S111, 200. L’Abbe, E. N., Kenyhercz, M., Stull, K. E., Keough, N., & Nawrocki, S. (2013). Application of Fordisc 3.0 to explore diﬀerences among cranial of North American and South African Blacks and Whites. Journal of Forensic Sciences, 58, 1579–1583. Landers, J. (1990). Gracia Real de Santa Teresa de Mose: A free black town in Spanish colonial Florida. American Historical Review, 95, 9– 30. Leathers, A., Edwards, J., & Armelagos, G. J. (2002). Assessment of classiﬁcation of crania using Fordisc 2.0: Nubian X-group test. American Journal of Physical Anthropology, S34, 99–100. Lewis, J. E., DeGusta, D., Meyer, M. R., Monge, J. M., Mann, A. E., & Holloway, R. L. (2011). The mismeasure of science: Stephen Jay Gould versus Samuel George Morton on skulls and bias. PLoS Biology, 9, e1001071. Lieberman, L. (2001). How “Caucasoids” got such big crania and why they shrank. From Morton to Rushton. Current Anthropology, 42, 69– 95. Littleﬁeld, D. (1977). Africans and Seminoles: From removal to emancipation. Westport, CT: Greenwood Publishing Group. Lorde, A. (1984). Sister outsider: Essays and speeches. Berkeley, CA: Crossing Press. Manica, A., Amos, W., Balloux, F., & Hanihara, T. (2007). The eﬀect of ancient population bottlenecks on human phenotypic variation. Nature, 448, 346–349. Marks, J. (2011). Plotz biology. Anthropomics. Available at: http://anthropomics.blogspot.com/2011/06/plotz-biology.html, accessed 28 April 2014. Martin, R. (1928). Lehrbuch der Anthropologie in systematischer Darstellung mit besonderer Ber€ ucksichtigung der anthropologischen Methoden. Band 2: Kraniologie, Osteologie. Jena: Gustav Fishcer. Meindl, R. S., & Lovejoy, C. O. (1985). Ectocranial suture closure: A revised method for the determination of skeletal age at death based on the lateral-anterior sutures. American Journal of Physical Anthropology, 68, 57–66. Jantz, R. L., & Logan, M. H. (2010). Why does head form change in children of immigrants? A reappraisal. American Journal of Human Biology, 22, 702–707. Michael, J. S. (1988). A new look at Morton’s craniological research. Current Anthropology, 29, 349–354. Jantz, R. L., & Meadows Jantz, L. (2000). Secular change in craniofacial morphology. American Journal of Human Biology, 12, 327–338. Miles, A. E. W. (2001). The Miles method of assessing age from tooth wear revisited. Journal of Archaeological Science, 28, 973–982. GELLER AND STOJANOWKSI | 17 Missall, J., & Missall, M. (2004). The Seminole wars: America’s longest Indian conﬂict. Gainesville, FL: University Press of Florida. Relethford, J. H. (2004). Global patterns of isolation by distance based on genetic and morphological data. Human Biology, 76, 499–513. Mock, S. B. (2012). Dreaming with the ancestors: Black Seminole women in Texas and Mexico. Norman: University of Oklahoma Press. Relethford, J. H. (2009). Race and global patterns of phenotypic variation. American Journal of Physical Anthropology, 139, 16–22. Mohanty, C. T. (1988). Under Western eyes: Feminist scholarship and colonial discourses. Feminist Review, 30, 61–88. Renschler, E. (2007). An osteobiography of an African diasporic skeletal sample: Integrating skeletal and historical information. PhD Thesis. Department of Anthropology, University of Pennsylvania. Montagu, A. (1997). Man’s most dangerous myth: The fallacy of race. Walnut Creek, CA: AltaMira Press. Montgomery, W. (1839). Remarks of Mr. Montgomery. In F. P. Blair, & J. C. Rives (Eds.), The congressional globe, containing sketches of the debates and proceedings of the twenty-ﬁfth congress. (pp. 268–270) Washington, DC: The Globe Oﬃce. Morton, S. G. (1839). Crania Americana: Or a comparative view of the skulls of various aboriginal nations of North and South America. Philadelphia, PA: J. Dobson. Morton, S. G. (1847). Hybridity in animals, considered in reference to the question of the unity of the human species. The American Journal of Science and Arts 3, 39–50, 203-212. Morton, S. G. (1849). Catalogue of skulls of man and the inferior animals, in the collection of Samuel George Morton. Philadelphia: Merrihew & Thompson, Prtrs. Morton, S. G. (1851). April 22nd. Proceedings of the Academy of Natural Sciences of Philadelphia, 5, 173–175. Motte, J. R. (1953). Journey into wilderness: An army surgeon’s account of life in camp and ﬁeld during the creek and Seminole wars, 1836-1838. Gainesville, FL: University of Florida Press. Mukhopadhyay, C. C., & Moses, Y. T. (1997). Reestablishing “race” in anthropological discourse. American Anthropologist, 99, 517–533. Mulroy, K. (2007). The Seminole freedmen: A history. Norman: University of Oklahoma Press. Ousley, S., D., Billeck, W. T., & Hollinger, R. E. (2005). Federal repatriation legislation and the role of physical anthropology in repatriation. Yearbook of Physical Anthropology, 48, 2–32. Ousley, S. D., & Jantz, R. L. (1998). The Forensic Data Bank: Documenting skeletal trends in the United States. In K. J. Reichs (Ed.), Forensic osteology. Advances in the identiﬁcation of human remains (pp. 441– 458) Springﬁeld, IL: Charles C. Thomas. Ousley, S. D., & Jantz, R. L. (2005). FORDISC 3.0 personal computer forensic discriminant functions. Knoxville: The University of Tennessee. Ousley, S. D., & Jantz, R. L. (2012). Fordisc 3 and statistical methods for estimating sex and ancestry. In D. Dirkmaat (Ed.), A companion to forensic anthropology (pp. 311–329) Malden, MA: Wiley-Blackwell. Ousley, S. D., Jantz, R. L., & Freid, D. (2009). Understanding race and human variation. American Journal of Physical Anthropology, 139, 68– 76. Pensley, D. S. (2005). The Native American Graves Protection and Repatriation Act (1990): Where the native voice is missing. Wicazo Sa Rev, 20, 37–64. Piper, H. M., Hardin, K., W., & Piper, J. G. (1982). Cultural responses to stress: Patterns observed in American Indian burials of the Second Seminole War. Southeastern Archaeology, 1, 122–137. Porter, K. W. (1996). The Black Seminoles: History of a freedom-seeking people. Gainesville, FL: University Press of Florida Rankin-Hill, L. (1997). A biohistory of 19th-century Americans: The burial remains of a Philadelphia cemetery. Westport, CT: Bergin & Garvey. Riordan, P. (1996). Finding freedom in Florida: Native peoples, African Americans, and colonists, 1670-1816. Florida Historical Quarterly, 75, 24–43. Rose, J. C. (Ed.). (1985). Gone to A better land: A biohistory of a rural black cemetery in the post-reconstruction south. Fayettesville, AR: Arkansas Archeological Survey. Rose, J. C., Green, T. J., & Green, V., D. (1996). NAGPRA is forever: Osteology and the repatriation of skeletons. Annual Review of Anthropology, 25, 81–103. Roseman, C. C. (2014). Troublesome reﬂection: Racism as the blind spot in the scientiﬁc critique of race. Human Biology, 86, 233–240. Roseman, C. C., & Weaver, T. D. (2004). Multivariate apportionment of global human craniometric diversity. American Journal of Physical Anthropology, 125, 257–263. Ross, A. H. (1999). Allocation of crania to groups via the ‘new morphometry’. Journal of Forensic Sciences, 44, 584–587. Ross, A. H., Slice, D. E., Ubelaker, D. H., & Falsetti, A. B. (2004). Population aﬃnities of 19th century Cuban crania: Implications for identiﬁcation criteria in south Florida Cuban Americans. Journal of Forensic Sciences, 49, 1–6. Ross, A. H., Ubelaker, D. H., & Kimmerle, E. H. (2011). Implications of dimorphism, population variation, and secular change in estimating population aﬃnity in the Iberian peninsula. Forensic Science International, 206, 214.e1–214.e5. Sattler, R. A. (1996). Remnants, renegades, and runaways: Seminole ethnogenesis reconsidered. In J. D. Hill (Ed.), History, power, and identity: Ethnogenesis in the Americas, 1492-1992 (pp. 36–69) Iowa City: University of Iowa Press. Sauer, N. J. (1992). Forensic anthropology and the concept of race: If races don’t exist, why are forensic anthropologists so good at identifying them? Social Sciences & Medicine, 34, 107–111. Sauer, N. J. (1993). Applied anthropology and the concept of race: A legacy of Linnaeus. Annals of Anthropological Practice, 13, 79–84. Schafer, D. L. (1993). “A class of people neither freemen nor slaves”: From Spanish to American race relations in Florida, 1821-1861. Journal of Social History, 26, 587–609. Sejrsen, B., Lynnerup, N., & Hejmadi, M. (2005). An historical skull collection and its use in forensic odontology and anthropology. Journal of Forensic Odonto-Stomatology, 23, 40–44. Sierp, I., & Henneberg, M. (2015). Can ancestry be consistently determined from the skeleton? Anthropological Review, 78, 21–31. Simmons, W. H. (1822). Notices of East Florida, with an account of the Seminole nation of Indians. Charleston, SC: A. E. Miller. Sparks, C. S., & Jantz, R. L. (2002). A reassessment of human cranial plasticity: Boas revisited. Proceedings of the National Academy of Sciences, 99, 14636–14639. Redman, S. J. (2016). Bone rooms: From scientiﬁc racism to human prehistory in museums. Cambridge, MA: Harvard University Press. Sparks, C. S., & Jantz, R. L. (2003). Changing times, changing faces: Franz Boas’s immigrant study in modern perspective. American Anthropologist, 105, 333–337. Relethford, J. H. (1994). Craniometric variation among modern human populations. American Journal of Physical Anthropology, 95, 53–62. Spradley, M. K., & Jantz, R. L. (2016). Ancestry estimation in forensic anthropology: Geometric morphometric versus standard and 18 | nonstandard interlandmark distances. Journal of Forensic Sciences, 61, 892–897. Stanton, W. (1960). The leopard’s spots: Scientiﬁc attitudes toward race in America, 1815-1859. Chicago: University of Chicago Press. Stocking, G. W. (1982). Race, culture, and evolution: Essays in the history of anthropology. Chicago: University of Chicago Press. Stojanowski, C. M. (2005). The bioarchaeology of identity in Spanish colonial Florida: Social and evolutionary transformation before, during, and after demographic collapse. American Anthropologist, 107, 417–431. Stojanowski, C. M. (2010). Bioarchaeology of ethnogenesis in the colonial southeast. Gainesville, FL: The University Press of Florida. Stojanowski, C. M. (2013). Mission cemeteries, mission peoples: Historical and evolutionary dimensions of intracemetery bioarchaeology in Spanish Florida. Gainesville, FL: The University Press of Florida. Stojanowski, C. M., & Duncan, W. N. (2009). Historiography and forensic analysis of the Fort King George “skull”: Craniometric assessment using the speciﬁc population approach. American Journal of Physical Anthropology, 140, 275–289. Strauss, A., & Hubbe, M. (2010). Craniometric similarities within and between human populations in comparison with neutral genetic data. Human Biology, 82, 315–330. Sturtevant, W. C., & Cattelino, J. (2004). Florida Seminole and Miccosukee. Handbook of North American Indians (Southeast), 14, 429–449. Sutphin, R., Ross, A. H., & Jantz, R. L. (2014). Secular trends in Cherokee cranial morphology: Eastern vs western bands. Annals of Human Biology, 41, 511–517. Tattersall, I. (2013). Stephen J. Gould’s intellectual legacy to anthropology. In G. Danieli, A. Minelli, & T. Pievani (Eds.), Stephen J. Gould: The scientiﬁc legacy (pp. 115–127). New York: Springer. Thomas, D. H. (2000). Skull wars: Kennewick man, archaeology, and the battle for Native American identity. New York: Basic Books. GELLER AND STOJANOWKSI Visweswaran, K. (1993). Predicaments of the hyphen. In Women of South Asian Descent Collective (Ed.), Our feet walk the sky: Women of the South Asian Diaspora (pp. 301–312). San Francisco: Aunt Lute Books. von Cramon-Taubadel, N. (2014). Evolutionary insights into global patterns of human cranial diversity: Population history, climatic and dietary eﬀects. Journal of Archaeological Science, 92, 43–77. Watkins, R. (2012). Variation in health and socioeconomic status within the W. Montague Cobb skeletal collection: Degenerative joint disease, trauma and cause of death. International Journal of Osteoarchaeology, 22, 22–44. Weik, T. M. (1997). The archaeology of maroon societies in the Americas: Resistance, cultural continuity and transformation in the African diaspora. Historical Archaeology, 31, 81–92. Weik, T. M. (2007). Allies, enemies and kin in the African-Seminole communities of Florida: Archaeology at Pilaklikaha. In T. Falola, A. Ogundiran (Eds.), Archaeology of Atlantic Africa and the African diaspora (pp. 311–331). Bloomington: Indiana University Press. Weik, T. M. (2009). The role of ethnogenesis and organization in the development of African-Native American settlements: An African Seminole model. International Journal of Historical Archaeology, 13, 206–238. Weisberg, M. (2014). Remeasuring man. Evolution & Development, 16, 166–178. Wescott, D. J., & Jantz, R. L. (2005). Assessing craniofacial secular change in American Blacks and Whites using geometric morphometry. In D. E. Slice (Ed.), Modern morphometrics in physical anthropology (pp. 231–245). New York: Springer. Wickman, P. R. (1999). The tree that bends: Discourse, power and the surkî peoples. Tuscaloosa, AL: University of Alabama vival of the Masko Press. Thomas, D. H., South, S., & Larsen, C. S. (1977). Rich man, poor men: Observations on three antebellum burials from the Georgia coast (Vol. 54). New York: Anthropological Papers of the American Museum of Natural History. Williams, F. L., Belcher, R. L., & Armelagos, G. J. (2005). Forensic misclassiﬁcation of ancient Nubian crania: Implications for assumptions about human variation. Current Anthropology, 46, 340–346. Tise, M. L., Kimmerle, E. H., & Spradley, M. K. (2014). Craniometric variation of diverse populations in Florida: Identiﬁcation challenges within a border state. Annals of Anthropological Practice, 38, 111–123. € ller, D., Vohberger, M., LutzWittwer-Backofen, U., Kästner, M., Mo Bonengel, S., & Speck, D. (2014). Ambiguous provenance: Experience with provenance analysis of human remains from Namibia in the Alexander Ecker collection. Anthropologischer Anzeiger, 71, 65–86. Trouillot, M. (1995). Silencing the past: Power and the production of history. Boston, MA: Beacon Press. Twyman, B. E. (1999). The Black Seminole legacy and North American politics, 1693-1845. Washington, D.C.: Howard University Press. Wright, R. (1992). Correlation between cranial form and geography in Homo sapiens: CranID - A computer program for forensic and other applications. Archaeology in Oceania, 27, 128–134. Ubelaker, D. H., Ross, A. H., & Graver, S. M. (2002). Application of forensic discriminant functions to a Spanish cranial sample. Forensic Science Communications, 4, 1–5. Wright, R. (2008). Detection of likely ancestry using CRANID. In M. Oxenham (Ed.), Forensic approaches to death, disaster and abuse (pp. 111–122). Sydney: Australian Academic Press. Urbanova, P., Ross, A. H., Jurda, M., & Nogueira, M. -I. (2014). Testing the reliability of software tools in sex and ancestry estimation in a multi-ancestral Brazilian sample. Legal Medicine, 16, 264–273. Wright, R. (2012). Guide to using the CRANID programs Cr6bInd: for linear and nearest neighbours discriminant analysis. Retrieved 11 October 2016 from http://osteoware.si.edu/comment/196.

RELATED PAPERS

RELATED TOPICS

Log In

The Vanishing Black Indian: Revisiting Craniometry and Historic Collections

The Vanishing Black Indian: Revisiting Craniometry and Historic Collections

Related Papers

RELATED PAPERS

RELATED TOPICS