The antebellum roots of distinctively black names

Abstract

This paper explores the existence of distinctively Black names in the antebellum era. Building on recent research that documents the existence of a national naming pattern for African American males in the late nineteenth and early twentieth centuries (Cook, Logan, and Parman, Explorations in Economic History 53:64–82, 2014), we analyze three distinct and novel antebellum data sources and uncover three stylized facts. First, the Black names identified by Cook, Logan and Parman using post-Civil War data are common names among Blacks before Emancipation. Second, these same Black names are racially distinctive in the antebellum period. Third, the racial distinctiveness of the names increases from the early 1800s to the time of the Civil War. Taken together, these facts provide support for the claim that Black naming patterns existed in the antebellum era and that racial distinctiveness in naming patterns was an established practice well before Emancipation. These findings further challenge the view that Black names are a product of twentieth century phenomena such as the Civil Rights Movement.

Keywords:

• Black names
• history
• demography
• Black family

JEL CLASSIFICATIONS:

• I1
• J1
• N3

Hold those things that tell your history and protect them. During slavery, who was able to read or write or keep anything? The ability to have somebody to tell your story to is so important. It says: ‘I was here. I may be sold tomorrow. But you know I was here’.

—Maya Angelou

1. Introduction

Given the unique social history of African Americans, it is remarkable that the historical development of racialized first names has received little scholarly attention. This is especially remarkable given the narrative literature’s focus on cultural development and persistence of culture among African Americans. Gutman (1976) notes that besides the studies of Puckett (1938, 1975) and Wood (1974), and observations from Mencken (1919), little has been written about African American naming systems in the American past. Even in this sparse literature, the majority of the scholarship has focused on naming patterns after Empancipation due to limits in data and methods to determine name distinctiveness before the Civil War. In using modern racially distinctive names, scholars have analyzed topics as wide ranging as labor market discrimination (Bertrand and Mullainathan 2004; Gaddis 2015), teacher and employer expectations (Gaddis 2017; Figlio 2005), and socioeconomic correlates of Black names (Fryer and Levitt 2004). However, this literature builds on the premise that distinctively Black names emerged as a product of the Black Power movement (Fryer and Levitt 2004), ignoring more historical relationships between Black identity, naming patterns and socioeconomic outcomes.

Cook, Logan, and Parman (2014) were the first to document an externally-validated national naming pattern among African Americans in the late nineteenth and early twentieth centuries. As they noted, “no historical narrative evidence we are aware of even suggests that such a robust, national naming pattern would exist…. The investigation of Black naming patterns links to the studies of the Black family, cultural development, and the postbellum development of African American identity” (p. 65). Names, they argue, are a part of the history of African Americans and also a key source for identifying the development of African American culture. The open question was whether the names they documented in the late nineteenth century had any precedence in the antebellum era, which would be indicative of persistent cultural developments with roots in the antebellem experience of enslavement.

This paper extends the reach of the names identified by Cook, Logan, and Parman as Black names to the pre-Civil War era. Exploiting three unique historical data soucres, and therefore three sources of external validation, we uncover three facts about Black names that add to our knowledge of racial naming patterns and, by extension, Black cultural development from enslavement to emancipation. First, we show that the Black names identified by Cook, Logan and Parman using post-Civil War data were common names among Blacks well before Emancipation. In particular, we find that roughly the same percentage of Blacks have Black names in the antebellum period as they did in the postbellum period. Second, these same Black names are racially distinctive in the antebellum period. Not only are these names relatively common among Blacks, but they are very uncommon among Whites at the time. Third, the racial distinctiveness of the names has a marked periodocity—distinctiveness increases from the early 1800s to the time of the Civil War. Using data that spans the nineteenth century, we find that the share of Blacks with Black names increases and, simultaneously, the share of Whites with these same names declines. By the eve of the Civil War, the racial naming pattern documented for the late nineteenth century was an entrenched feature in the United States.

This finding has important implications for racial naming patterns and their history. Scholars of African American cultural history have long held that the unique development of African American culture involves both familial and larger social ties among people from various ethnic groups in the African Diaspora. How this culture developed, synthesized, and persisted in a chattel slavery system with high levels of mobility via trading in enslaved individuals is a unique historical question. These results show that, as enslavement continued through the nineteenth century, African American culture included naming practices that were national in scope by the time of Emancipation. Since none of the names analyzed here are of African origin, the results here point to first names as a distinct African American cultural practice which began during enslavement in the United States.

2. History of Black names

Engerman (1978) notes that names play an important role in our understanding of African American social development, and yet they remain under-analyzed, a missing piece of the historical scholarship. The historical literature focused on whether naming conventions defy the conventional wisdom that the institution of slavery destroyed African American kinship patterns. Gutman (1976) spends considerable time documenting how naming patterns are consistent with strong kinship bonds among African Americans. One common practice was to name the eldest son after the father. Gutman’s analysis of the 1880 census revealed that nearly a quarter of African American households had a son named for his father. While Gutman's analysis suggests that a portion of African American families named sons after elder men in the family we have no additional evidence on persistence of the pattern.

Gutman argues that African Americans exhibited a great deal of control over the naming of their offspring in slavery, which is consistent with Blassingame's (1972), Wood's (1974) and Genovese's (1974) histories of plantation life. Cody (1982) argues that the naming of enslaved children by their parents was an important way of establishing their place in the slave community. First names could refer to parents, grandparents, and other elder members as a way of establishing familial links. There is no narrative evidence that names were related to slave occupations. In the absence of surname salience, first names of children appear to be prominent carriers of family history. However, the historical scholarship in the social sciences has not paid a great deal of attention to Black names with the exception of a few studies (such as Cassidy 1966; Puckett 1938; Puckett 1975; DeCamp 1967; Price and Price 1972). Indeed, Gutman is the most recent largescale analysis. Research in the humanities, however, is rich with names as descriptive carriers of historical legacies and also as exercises in power (Benston 1982; Cooke 1977; King 1990; Green 2002).

This is not to say that these conclusions are uncontested. While scholars have added greater nuance to the destructive nature of the slave economy on African American family bonds laid out by Frazier (1939), the actual stability of the family is an open question. Fogel (2003) questions the reliability of Gutman's evidence since it comes from large plantations where familial structure would be more likely to be intact and where enslavers would be least likely to be involved in the task of naming newborn enslaved children. This implies that the search for historical names would need to contend with differences in autonomy over naming that may be a function of plantation size itself. According to some research, for example, enslavers began disproportionately giving enslaved Africans biblical names as a way to force them away from the African day names that were commonly used in the 18th century and to encourage the adoption of Christianity (Inscoe 1983).

A related issue is the role of the slave trade in the development of naming conventions among African Americans more generally. The movement of enslaved people and continual high levels of mobility and family breakup may have necessitated and accelerated the development of broader cultural norms which would have helped African Amerians acclimate to new enviornments. Steckel and Ziebarth (2013), for example, find that slave trading was a significant fraction of western movement in the late antebellum era, and they argue that such trading was likely disruptive to the slave family. Kaye (2007), however, finds that slave neighborhoods in the western portions of the Cotton South created strong familial bonds. Both factors could have been reinforcing—name stability could reflect a desire to have some semblance of stability despite the high likelihood of breakups of community bonds.

In general, there is little historical consensus on enslaved naming patterns. Blassingame (1972) and Elkins (1959) differ on the weight that one should attach to naming patterns. Cassidy (1966) and DeCamp (1967) show that African naming conventions were perverted in the New World. For example, Sambo, a name which became associated with a shiftless individual, is actually a Hausa name for the second son (Wood 1974). Engerman (1978) notes that who named enslaved children, and whether enslavers retained veto power over names, is unknown. It could be the product of resistence (Stampp 1956), part of an implicit cultural norm (Genovese 1974), reflect greater community attachment that extended beyond specific plantations (Kaye 2007), or be a forced effort by enslavers to assimilate to Christianity (Inscoe 1983). Therefore, the study of enslaved names is more complicated than racial naming patterns alone, since it is unclear who named whom, and individual motivations behind naming patterns are difficult to discern from the historical record.

While acknowledging these issues would be present in any analysis, we explicitly avoid some of these issues by focusing on names identified postbellum, when Blacks had control over naming patterns. Given that Cook, Logan, and Parman have identified postbellum Black names, we ask whether those same names are (1) held by a large number of Blacks before Emancipation and (2) whether they were racially disproportionate in the antebellum era. To the extent that the postbellum names were common and racially disproportionate in the past, it provides evidence that the names identified are not a product of Empancipation, but rather a cultural norm developed in chattel slavery. Equally important, if the names identified post-bellum are disproportionately held by Blacks over a wide geographic area, it stands to reason that Black names were a national naming pattern among the enslaved, which points to a unique feature of cultural development among Black people in the United States that has not been empirically analyzed. It would also suggest that rather than destroying or amplifying familial bonds, naming practices could have been one way of establishing racial identify in a era where mobility and familial relations were subjected to outside forces.

In sum, there are advantages and disadvantages to using a set of names derived from the postbellum era. A disadvantage of this apporach is that it cannot identify names which were common in the antebellum era and which ceased to be racially distinctive thereafter. A key advantage, as noted above, is that they limit the degree to which these names would reflect restrictions on naming due to enslavement. The names used here were not assigned nor aproved of by Whites. Second, the names could reflect a change in naming practices post-bellum. As such, any evidence of these same names in the antebellum era would be strong evidence of persistence of names irrespective of how they were derived during the antebellum era.

3. The historical African American names

We obtain the set of historical Black names from Cook, Logan and Parman (2014), who adopt a transparent approach to identify Black names that begins with a geographically stratified sample of Black households in census data and then seek to (1) internally validate the distinctive nature of the names in census records and (2) use a wealth of new, broadly representative data from different sets of locations to verify the distinctiveness of the names in the census records. Their methodological approach builds on the conjecture that if Black naming patterns existed in the past they would be found in independent samples of the Black population, and the relative distinctiveness of the names would be highly correlated in those independent samples. They begin by constructing an index of African American names for the period 1900 to 1920 drawn from the District of Columbia and three states: Georgia, Michigan and New York. This was done to geographically stratify the naming pattern to account for the population distribution of the Black population at end of the nineteenth century. These names were verified in the entire census and further verified in three independent data sources with broad historical coverage. Subsequent to publucation of their original study, the distinctive nature of the names was confirmed in yet another independent historical data source (Cook, Logan, and Parman 2016). In sum, they find a striking, robust degree of distinctiveness for small set of African American names. They conclude that the national naming pattern was a hitherto unknown fact in African American culture, and is the first and only national naming pattern in the literature for this time period. The names are listed in Table 1, and we use those names in this analysis.¹

Table 1. Historical African American first names.

Abe

Abraham

Alonzo

Ambrose

Booker

Elijah

Freeman

Isaac

Isaiah

Israel

King

Master

Moses

Pearlie

Percy

Perlie

Purlie

Presley

Presly

Prince

Titus

Source: Cook, Logan, and Parman (2014).

It is important to note that using terms such as “Black names” can result in confusion since the names themselves are not of African origin. Indeed, given the geographic scope of the African Diaspora, the number of names which can be considered “Black” would be large, covering the African continent, the Caribbean, and North, Central, and South America. Even today, contemporary names that are referred to and used as examples of “Black names” in empirical research do not have African origins. For example, the racially distinct names used by Bertrand and Mullainathan (2004), whose celebrated study found negative labor market impacts due to inferred race on resumes, were not of African descent. Their analysis included names such as Jermaine (of Latin/French origin), Darnell (Anglo Saxon), Tyrone (Irish), and Leroy (Norman), and are routinely used and known as Black names despite their European origin. Lieberson and Mikelson (1995) note that contemporary Black names such as Shaniquq and Lakisha share a linguistic stricture but not an African origin. While these names are recognized and described in the literature as “Black names,” this is a term that refers to usage and not origin. In fact, referring to them as Black names, which explicitly notes their contemporaneous racial connotation, divorces them from an ethnic explication. Any Black name of African origin would not be known by or referred to as a Black name, it would be identified by its ethnic origin.

Following the existing literature on racial naming patterns, we use terms such as “Black names,” “racial naming patterns,” and “distinctively Black names” as one of contextual usage and disproportionality, not of linguistic or ethnic orgin. It is important to note that names are designated as “Black” or “White” names in the literature are defined by their racial disproportionality and in context. Names infer race, gender, and socioeconomic position due to their disproportionate use by certain groups. We make no assumption that a name must be of African or European origin to become a “Black” or “White” name. As an example, names such as Ashley, Courtney, and Sydney are contemporarily thought of as women’s names given their gender disproportionality, but these were common given names for men in the past, and some men have these names today. To call them “women’s names” refers to their contemporary usage, which must be contextualized to the time. A similar analysis is applied to the names we refer to as Black names here.

The question we ask here about Black names speaks to the racial disproportionality used to define the term in the empirical literature. Given that the names found in Cook, Logan, and Parman (2014) were racially distinct in the late nineteenth and early twentieth century, does that hold for the antebellum past as well? That is, can we think of the historical Black names as being part of antebellum history as well? Despite their origins, were names such as Abraham and Isaac racially distinct before the Civil War? If so, were their rates of disproportionality similar to what Cook, Logan, and Parman found for the postbellum period? This is inherently and empirical question, where one must look at the rate of disproportionality of the names as used in antebellum America.

4. Data

We use three distinct data sources to document the antebellum naming practices among African Americans. Importantly, two of our data sources also allow us to directly assess the racial distinctiveness of the names as they contain a large number of White names. We further supplement the evidence of racial name distinctiveness by analyzing White names in the census of 1850. We describe each data source below.

4.1. Louisiana Hall data

The Louisiana Hall data are a compilation of information on over 100,000 enslaved people who came to Louisiana between 1719 and 1820. The underlying sources for the Hall data are diffuse—they come from sales records, newspaper listings, and international archives. The data were first published in the volume edited by Gwendolyn Midlo Hall (2000) “Databases for the Study of Afro-Louisiana History and Genealogy 1699–1860.” For our purpoes, the data are especially important as Louisiana, for historical reasons, may display names less consistent with a national naming pattern as other data sources. Given the preponderance of French culture, we would expect fewer of the names identified post-bellum in Cook, Logan and Parman, and which are national in nature, to be prominent in Louisiana at the time. Another strength of the data is that they are longitudinal, allowing us to see how or if Black naming patterns changed over time. One drawbaack of the data is that they contain only the names of the enslaved—it is not possible to compare the names in the data to a contemporatneous sample of Whites using the same methodology.

4.2. New Orleans slave sales data

The New Orleans slave sales data were compiled by Pritchett and described in Calomiris and Pritchett (2016). Those authors collected data on over 16,000 slave sales between 1856 and 1861. Unlike states with a common law tradition, Louisiana treated the enslaved like real estate, and slave sales had to be recorded and notarized in order to establish title (Louisiana 1806, section 10). Today, the records of many of these slave sales may be found in the New Orleans Notarial Archives and the New Orleans Conveyance Office. Because of the availability of these records and the size of the market, New Orleans is the best source for data on slave sales within the United States. Since sales were considered property transactions in Louisiana, the sales data here list the names of both buyers, sellers, and the enslaved. Also, the recorded age of the enslaved is retained as well.

During this time, New Orleans was the largest city in the South and the site of its largest slave market. Indeed, as Calomiris and Pritchett (2016) argue, this data represents a “national” sample as traders used the New Orleans market as a national clearinghouse for enslaved individuals. Unlike the Hall data, the New Orleans sales data are not confined to enslaved persons from Louisiana, as these enslaved individuals were transported to and from New Orleans as part of the national slave trade. Another advantage of the New Orleans data is that the names of purchasers and sellers are recorded, and this allows us to contemporaneously investigate whether the names of the enslaved matched the names of those buying and selling in the market.²

4.3. Coastwise Manifest data

The third data source we use are the names from Coastwise Manifest of shipments of enslaved individuals. While shipment of enslaved individuals was common before the Revolutionary War, the Act of 1807, signed into law by President Thomas Jefferson, made the international slave trade illegal. The domestic slave trade was still legal and because of this distinction slave manifest records became more detailed as to ensure that imported enslaved individuals were not being transported domestically. Before 1808, the information on the enslaved on board these ships was less informational and sometimes was solely the number of enslaved people onboard. Post-1808, personal information for each enslaved person was included and the owner had to affirm the enslaved had been imported prior to 1808. As such, the manifests records we use include information on name of ship, master, port of departure, port of destination, list of enslaved on board, in addition to the first name, sex, age and stature of the enslaved. These records have been used to investigate enslaved health in the past (see Margo and Steckel 1982), but they have never been used for name analysis. These data have three key strenghs. First, the data are longitudinal and therefore we can analyze names over time. Another strength is that the names of owners are recorded, which allows for contemporaneous comparisons. Third, the manifest records are from disparate sources covering a relatively wide geographic area. As such, they are not simply a regional sample of names but closer to a national sample given the wide range of transportation in enslaved people at the time (see Steckel and Ziebarth 2013).

5. Antebellum Black names

5.1. The popularity of Black names in the antebellum era

We begin the discussion of names with the Louisiana Hall Data. Table 2 shows the number of men in the Hall data who have each of the names identified in Cook, Logan, and Parman. Overall, slightly more than 1.25% of the 33,774 men with names given in the Hall Data have a historical Black name. In Table 3, the same information is displayed for the New Orleans Sales Data. There, more than 3.75% of the enslaved men in the sales data have one of the historical Black names. This percentage is greater than the percentage of Black men holding historical Black names reported in Cook, Logan, and Parman (2014), and is suggestive of Black names being even more common in the antebellum era. Table 4 shows the pattern for the coastwise manifest data containing names for just over 10,000 enslaved men. There, we see that over 3.5% of all enslaved men had one of the historical Black names identified with post-bellum data.

Table 2. Black name disproportionality in Hall Louisiana Data.

Count of all males with given first name
First name	Enslaved
Abe/Abraham	99
Alonzo	1
Ambrose	29
Booker	2
Elijah	11
Freeman	1
Isaac	214
Isaiah	2
Israel	3
King	4
Master	0
Moses	120
Percy	1
Perlie/Purlie/Pearlie	0
Presley/Presly	0
Prince	126
Titus	13
	626
Share who have a Black name	1.36%

Note: Females, unnamed, and names with initials only are excluded when calculating share of individuals with a Black name.

Table 3. Black name disproportionality in New Orleans Sales Data.

Count of all males with given first name
First name	Enslaved	Buyers	Sellers
Abe/Abraham	49	33	17
Alonzo	9	6	1
Ambrose	8	19	8
Booker	1	0	0
Elijah	9	4	8
Freeman	0	0	0
Isaac	85	40	37
Isaiah	13	2	0
Israel	5	8	5
King	3	0	0
Master	0	0	2
Moses	77	24	16
Percy	0	0	0
Perlie/Purlie/Pearlie	1	0	0
Presley/Presly	2	0	0
Prince	15	0	0
Titus	4	0	0
	281	136	94
Share who have a Black name	3.78%	0.89%	0.68%
Disporportionality		4.26	5.55

Note: Female names and names with initials only are excluded when calculating share of individuals with a Black name.

Table 4. Black name disproportionality in Coastwise Manifest Data.

Count of all males with given first name
First name	Enslaved	Owners
Abe/Abraham	61	7
Alonzo	2	0
Ambrose	0	0
Booker	2	0
Elijah	6	0
Freeman	0	1
Isaac	165	49
Isaiah	9	1
Israel	8	0
King	6	5
Master	0	0
Moses	65	20
Percy	1	0
Perlie/Purlie/Pearlie	0	0
Presley/Presly	0	1
Prince	56	0
Titus	18	0
	399	84
Share who have a Black name	3.65%	0.39%
Disporportionality		9.31

Note: Female names and names with initials only are excluded when calculating share of individuals with a Black name.

The results point to two new facts about historical Black names. First, Black names were quite common among Blacks as measured in contemporaneous data. Indeed, the proportions in two of the data sources show a higher fraction of Blacks having Black names than what was found in the late 19th century data. Second, two of the data sources (the sales data and the coastwise manifest data) are national in scope and provide evidence that the national post-bellum naming pattern which existed after the Civil War was, in fact, a continuing phenomenon.

Although we are agnostic about the individual names (we reiterate that our goal is to see how prevalent the names are as a group), it is important to note that not all of the names identified in Cook, Logan, and Parman are found in the antebellum sources. For example, Master, Presley, and Freeman are rarely seen for enslaved men. Other names, such as Isaac, Abraham, and Moses, are the most common of the Black names found here. This trend implies that some names are likely to have appeared post-Emancipation, and in the case of names such as Master and Freeman, the reasons would be obvious.

5.2. Racial name distinctiveness in the antebellum era

Returning to Table 3, we use the names of buyers and sellers in the New Orleans Data to see how racially distinctive the Black names were at the time. As the results show, fewer than 1% of buyers or sellers in the New Orleans market had Black names. Enslaved individuals were more than 4 times as likely as buyers to have a Black name. Enslaved individuals were more than 5 times as likely as sellers to have a Black name. In Table 4, a similar exercise is performed with the coastwise manifest data which contains names for just over 20,000 slave owners. There, we see that fewer than 0.5% of listed owners are likely to have a Black name, and enslaved individuals are more than 9 times as likely to have a Black name than an enslaver in the data.

One concern would be that name distinctiveness via comparison to Whites in the sales and manifest data would make a poor comparison group to show racial name distinctiveness. While it should be noted that the direct comparison with names recorded at the same time is illuminative, it could be that Blacks held names that were uncommon among their White enslavers but not in general. A further comparison, provided in Table 5, can be made with all Southern Whites from the census data for 1850 and 1860. While over 3.5% of Blacks in sales markets or manifest records had Black names, Whites held these names at less than half of that rate. Among White children, the relationship is even less, with fewer than 1.25% holding a Black name. At baseline, then, Black names were, at a minimum, twice as common among the enslaved as they were among all Southern Whites. Overall, the results provide strong suggestive evidence that Black names were common among the enslaved and were quite racially disproportionate in the antebellum era.

Table 5. Black name disproportionality for Whites in Census Data.

Count of all males with given first name
First name	All Whites 1860	White children 1850	All Whites 1850
Abe/Abraham	281	4	293
Alonzo	159	3	106
Ambrose	1	3	2
Booker	2	0	3
Elijah	250	8	209
Freeman	52	1	27
Isaac	1014	28	694
Isaiah	85	3	67
Israel	88	0	70
King	5	0	4
Master	0	0	0
Moses	284	2	232
Percy	11	1	0
Perlie/Purlie/Pearlie	3	1	5
Presley/Presly	14	0	11
Prince	4	1	3
Titus	14	0	3
	2267	55	1729
Share who have a Black name	1.68%	1.16%	1.74%

Note: Females, unnamed, and names with initials only are excluded when calculating share of individuals with a Black name.

5.3. The increasing concentration of Black names in the antebellum era

Some of the data sources we use for antebellum names allow us to uncover a time-pattern of the names, to see when and how the names increased in prevalence among the enslaved. In Table 6, we decompose the Hall data, which is from Louisiana, by birth cohort. The time pattern in Table 6 is striking—none of those in the Hall data born before 1700 had a Black name, but for those born between 1780 and 1800 nearly 2% had a Black name. Recalling that in the data overall around 1.25% of men had Black names, this time trend is striking. Indeed, one of the reasons that Black names were relatively uncommon in the Hall data relative to the other sources is that the data contains such a large number of enslaved individuals born before 1740, when less than one half of one percent of those in the data held Black names. By the turn of the 19th century, Black names in Louisiana were nearly as common as they were in other data sources.

Table 6. Name distinctiveness over time from Hall Louisiana Data.

Above age 9	With distinctive name
Birth year	Number	Percentage	N
Before 1700	0	0.00%	125
1700–1720	3	0.60%	503
1720–1740	2	0.13%	1512
1740–1760	39	0.84%	4633
1760–1780	81	0.82%	9892
1780–1800	267	1.94%	13,757
1800–1820	67	2.14%	3124

Notes: Excludes those for whom birth year cannot be determined.

For males above age 9 in the data

We show the same exercise for the coastwise manifest data in Table 7. The trend over time is even more striking than for the Hall data. For those born 1770 to 1790, 3.17% were likely to hold a Black name, but for the 1810–1830 birth cohort more than 4.5% of the enslaved men held a Black name. Taken together, Tables 6 and 7 imply a time trend in the prevalence of Black names that implies that Black names increased over time.

Table 7. Name distinctiveness over time from Coastwise Manifests.

	With distinctive name
Birth year	Number	Percentage	N
Before 1770	0	0.00%	0
1770–1790	6	3.17%	189
1790–1810	27	4.06%	665
1810–1830	79	4.65%	1698

Notes: Excludes those for whom birth year cannot be determined.

In Table 8, we do the same for Whites using the 1850 Census data for Southerners. The trend in Black names is the exact opposite of that in Tables 6 and 7 and suggests a strong decline in Black names among Whites. For Whites born before 1770, more than 4.75% held Black names, but for the 1810–1830 birth cohort, less than 2% held Black names. Table 8 also shows results for the 1860 Census and finds the same pattern. For Whites born between 1790–1810, more than 2.55% held Black names, but for the 1810–1830 birth cohort, less than 2% held Black names. As we noted earlier, this trend obscures the fact that, among the youngest cohorts, Black names were relatively rarely held among Whites.

Table 8. Name distinctiveness over time for Whites in 1850 and 1860 Census Data.

	1850 census			1860 census
Birth year	Number	Percentage	N	Number	Percentage	N
Before 1770	13	4.87%	267
1770–1790	105	3.24%	3242	6	1.85%	325
1790–1810	289	2.36%	12,239	54	2.55%	2116
1810–1830	567	1.84%	30,760	107	1.81%	5915

Notes: Excludes those for whom birth year cannot be determined.

6. Discussion and future directions

Finding the presence of an antebellum Black naming pattern is a novel advance in the quantitative history of Afrian American culture. Since the earliest histories of African American kinship (Frazier 1930), scholars have investigated ways in which Black culture formed and persisted via family bonds. Names are likely one of those forms of cultural transmission. Names could have taken on an even more pronounced role given the high rate of family breakup caused by the slave trade (Logan and Pritchett 2018 and others). Given this, we believe that other data sources could be used to further validate the Black names. These would include advertisements for runaway slaves (which would also include names of enslavers), information in other administrative data such as probate records, and plantation sources which note enslaved picking rates. Although each of these sources contain some elements of bias, new work exploiting large antebellum data sources could extend the results here.

Similarly, the relationship between the slave trade and the development of the national pattern uncovered here merits further investgation. Both of the data sources more closely linked to the transportation of enslaved people point to a high degree of name distinctiveness. Is such a pattern a product of the slave trade itself or did the slave trade involve the renaming of enslaved individuals, which persisted as familial bonds after emancipation? More detailed and archival research is needed to answer the question of transfers of ownership, westward movement, and naming practices in the antebellum era.

7. Conclusion

This paper presented the first evidence of racially distinctive naming patterns among African Americans in the antebellum era. Indeed, the time pattern of names suggest that Black names increased in their racial disproportion as the antebellum period came to an end. This was not only the result of younger Blacks being more likely to have Black names, but also from Whites who were less likely to have Black names. The racial separation of names is an intriguing finding that has implications for how racial naming patterns came about and how the actions of both Whites and Blacks fed into that process. To the extent that historians have argued that names in the antebellum era were an attempt to strip Blacks from their African heritage (Inscoe 1983), it is unclear why Whites would move away dramatically from these names over the antebellum period. This is all the more surprising since the names analyzed here have only recently been discovered, and yet it appears as a regularity in the data as a pattern for both Black and White names.

As noted by Cook, Logan, and Parman (2014), the persistence of Black names, and the trend in naming patterns over time shown here, should renew calls for quantitative history to retain more traditional methods of analysis. Our method and results harken back to an earlier style of quantitative historical scholarship which produced new facts that altered our understanding of the past. Fogel (1975a: 337) notes that such methods form the backbone of rigorous quantitative historical work: "The most common method of direct measurement in history is counting. My reference to counting as a rigorous method of measurement is not to be taken derisively. I use portentous language for what appears to be an elementary operation partly because I want to emphasize the dramatic change in interpretation that may result merely by moving from an impression to an actual count." As with the original research on Black names, this paper provides such a dramatic reinterpretation. Black names were clearly evident during the antebellum era, but new names appeared as Black names after. Whites appear to have been moving away from such names as well. Neither of these facts relies on overly sophisticated methods, but rather a straightforward desire to describe patterns in the data.

The antebellum history of Black names opens up a large number of new questions. Can we identify from the narrative record why these names came to be so common as opposed to others? Were Whites and Blacks aware of the stark disproportionality of these names, and did this influence naming patterns in ways we can see in the historical record? Are different types of Blacks (say, those enslaved on large plantations) more likely to have Black names? Are Whites who enslaved Blacks more or less likely to have Black names compared with other Whites? Each of these new questions can now move to a fuller analysis of the role of enslavement in the development of naming practices.

Notes

1 Unfortunately, it is not possible to perform a similar exercise for women’s names given the Southern convention of first and middle names (e.g., Sarah Jane, Mary Agnes, Mary Ann). We find much lower baseline distinctiveness using women’s names. See Cook, Logan, and Parman [2014] for additional details.

2 Of particular note is that fact that traders are noted separately from buyers and sellers.