USF Libraries
USF Digital Collections

Nearer, my farm, to thee :


Material Information

Nearer, my farm, to thee : a spatial analysis of african american settlement patterns in hillsborough county, florida
Physical Description:
O'Brien, Matthew Andrew
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Southeastern U.s
Dissertations, Academic -- Archaeology -- Masters -- USF   ( lcsh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


ABSTRACT: Geographic Information Systems (GIS) have demonstrated their utility in predictively modeling the location of archaeological sites, and providing a framework for cataloging sites eligible for heritage management status. The intent of this GIS-based study is to begin to create a geohistorically organized database of information culled from historic documents and archaeological excavation. In this case study of postbellum land tenure in Hillsborough County, Florida, a GIS-based approach is used to demonstrate the impacts of federal and state land ownership policy decisions during the Reconstruction Era and beyond. GIS data are also used to reveal information about how people use their allotted environment to non-verbally communicate their perceptions of the world and their place in it. Finally, GIS are shown to be ideally suited for allowing multi-scalar, diachronic comparisons of archaeological sites and materials. This research was conducted according to the concepts of Actor-Network-Theory (ANT), which assumes there is a generalized symmetry between the agency of human actors and non-human actants (i.e. it does not assume the primacy of human intentional action). ANT accepts that materials can carry non-verbal messages (e.g. colors, aromas, tactility), which affect how humans interact, communicate, and organize themselves in space. ANT allows for the use of scales based on human action, and analyses that are based standardized metrologies. Finally, ANT obviates being limited to strict categories of macro- and micro-, by accepting that networks may bridge both. This research shows that two rural communities have undergone similar growth trajectories, with a historically black community having experienced some setbacks in the early 20th century. However, the results show that the rural African American community was not more subdivided than the neighboring Euro-American community, contrary to initial expectations. Additionally, there is a suggestion that communities may move socially important buildings such as churches schools to the community center or periphery, depending on the intended recipient of the message. The study also documents the centralization, concentration, and clustering of the county's African American population through time.
Thesis (M.A.)--University of South Florida, 2011.
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Matthew Andrew O'Brien.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 153 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0004927
usfldc handle - e14.4927
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2011 flu ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004927
XX9999 (Online)
1 100
O'Brien, Matthew Andrew.
0 245
Nearer, my farm, to thee :
h [electronic resource] /
b a spatial analysis of african american settlement patterns in hillsborough county, florida
by Matthew Andrew O'Brien.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 153 pages.
(M.A.)--University of South Florida, 2011.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Geographic Information Systems (GIS) have demonstrated their utility in predictively modeling the location of archaeological sites, and providing a framework for cataloging sites eligible for heritage management status. The intent of this GIS-based study is to begin to create a geohistorically organized database of information culled from historic documents and archaeological excavation. In this case study of postbellum land tenure in Hillsborough County, Florida, a GIS-based approach is used to demonstrate the impacts of federal and state land ownership policy decisions during the Reconstruction Era and beyond. GIS data are also used to reveal information about how people use their allotted environment to non-verbally communicate their perceptions of the world and their place in it. Finally, GIS are shown to be ideally suited for allowing multi-scalar, diachronic comparisons of archaeological sites and materials. This research was conducted according to the concepts of Actor-Network-Theory (ANT), which assumes there is a generalized symmetry between the agency of human actors and non-human actants (i.e. it does not assume the primacy of human intentional action). ANT accepts that materials can carry non-verbal messages (e.g. colors, aromas, tactility), which affect how humans interact, communicate, and organize themselves in space. ANT allows for the use of scales based on human action, and analyses that are based standardized metrologies. Finally, ANT obviates being limited to strict categories of macro- and micro-, by accepting that networks may bridge both. This research shows that two rural communities have undergone similar growth trajectories, with a historically black community having experienced some setbacks in the early 20th century. However, the results show that the rural African American community was not more subdivided than the neighboring Euro-American community, contrary to initial expectations. Additionally, there is a suggestion that communities may move socially important buildings such as churches schools to the community center or periphery, depending on the intended recipient of the message. The study also documents the centralization, concentration, and clustering of the county's African American population through time.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Weisman, Brent R.
Southeastern U.s.
Dissertations, Academic
x Archaeology
t USF Electronic Theses and Dissertations.
4 856


Nearer, My Farm, to Thee: A Spatial Analysis of Afri can American Settlement Patterns in Hillsborough County, Florida by Matthew A. O’Brien A thesis submitted in partial fulfillment Of the requirements for the degree of Master of Arts Department of Anthropology College of Arts and Sciences University of South Florida Major Professor: Brent R. Weisman, Ph.D. Thomas J. Pluckhahn, Ph.D. Susan D. Greenbaum, Ph.D. Date of Approval: March 28, 2011 Keywords: Southeastern U.S., Geohistorical, GIS, Archaeo logy, ANT Copyright 2011, Matthew A. O’Brien


Dedication For the Dl gCais


Acknowledgments Sincere gratitude is owed, and hereby given, to the a dministrative staff of the USF Anthropology Department, the staff of the USF Li braries Special Collections Department, and the members of Bealsville, Inc.


i Table of Contents List of Tables ..................................... ................................................... .................ii List of Figures .................................... ................................................... ................ iii Abstract ............................................ ................................................... .................iv Chapter 1: Introduction ............................ ................................................... .......... 1 Chapter 2: Physical Setting ........................ ................................................... ..... 17 Chapter 3: Historic Background ........................ .................................................. 21 Postbellum Settlement Patterns in the American South ................................. 22 Historic Context for Using GLO Records in Florida ..... .................................... 26 Settlement Histories of Bealsville and Springhead .... ...................................... 28 Comparing Local Rural Sites with Other Areas of the A merican South .......... 34 Settlement History of Tampa’s Historically Black Neighbor hoods ................... 38 Settlement History of Willow Company Town .......... ....................................... 50 Chapter 4: Research Background ....................... ............................................... 55 Settlement Archaeology ............................. ................................................... .. 69 The Built Environment as Non-Verbal Communication ... ................................ 72 Material as Behavior ............................... ................................................... ..... 78 Research Design and Meta-Design ....................... ......................................... 82 Chapter 5: Research Methods ........................ ................................................... 88 Chapter 6: Results ................................ ................................................... ........... 94 Chapter 7: Discussion ................................. ................................................... .. 118 Chapter 8: Conclusion ............................... ................................................... .... 129 References Cited ................................... ................................................... ........ 133


ii List of Tables Table 1. Distance to Baptist Church in Bealsville ..... ........................................ 100 Table 2. Distance to Baptist Church in Springhead .... ...................................... 100 Table 3. Bealsville and Springhead Distances to Schoo ls ................................ 102 Table 4. Bealsville Structure to Structure Distances ... ...................................... 104 Table 5. Springhead Structure to Structure Distances ..................................... 104 Table 6. Property Dispositions of Bealsville Founders ..................................... 109 Table 7. Property Dispositions of Springhead Founders .................................. 110 Table 8. Robles Pond Distances to Community Structures ............................. 111 Table 9. Neighborhood Population Densities in Tampa (1927) ........................ 115 Table 10. Distances to Group Services for Willow ....... ..................................... 115


iii List of Figures Figure 1. GLO Map of T29S, R22E ................... ................................................... 2 Figure 2. Physiography of Hillsborough County, Florid a ..................................... 18 Figure 3. Tampa’s Historically Black Neighborhoods ..... .................................... 45 Figure 4. Planview of Willow, Florida ............. ................................................... .. 53 Figure 5. GLO Locations of Bealsville Founder Propert ies ................................. 95 Figure 6. Growth of Bealsville’s Built Environment .. ........................................... 96 Figure 7. GLO Locations of Springhead Founder Proper ties .............................. 98 Figure 8. Growth of Springhead’s Built Environment ......................................... 99 Figure 9. Soil Drainages for Bealsville and Springhea d .................................... 103 Figure 10. Comparison of Structure to Structure Distance s ............................. 105 Figure 11. Subdivision of Founder Properties Through 2008 ........................... 107 Figure 12. Robles Pond in 1922 and 1931 ............ ........................................... 112 Figure 13. Black Residential Centralization in Tampa (2000) ........................... 114 Figure 14. Black Residential Concentration in Hillsboro ugh County (2000) ..... 116 Figure 15. 1916 Dixie Survey Map Showing Property O wners ......................... 121


iv Abstract Geographic Information Systems (GIS) have demonstrated their utility in predictively modeling the location of archaeological si tes, and providing a framework for cataloging sites eligible for heritage ma nagement status. The intent of this GIS-based study is to begin to create a geohistorically organized database of information culled from historic documents a nd archaeological excavation. In this case study of postbellum land ten ure in Hillsborough County, Florida, a GIS-based approach is used to demonstrate the impacts of federal and state land ownership policy decisions during the Reconstru ction Era and beyond. GIS data are also used to reveal information about h ow people use their allotted environment to non-verbally communicate their percepti ons of the world and their place in it. Finally, GIS are shown to be ideally sui ted for allowing multi-scalar, diachronic comparisons of archaeological sites and materia ls. This research was conducted according to the concepts of Actor-NetworkTheory (ANT), which assumes there is a generalized symme try between the agency of human actors and non-human actants (i.e. it does not assume the primacy of human intentional action). ANT accepts that materials can carry nonverbal messages (e.g. colors, aromas, tactility), which aff ect how humans interact, communicate, and organize themselves in space. ANT allow s for the use of scales based on human action, and analyses that are based st andardized


v metrologies. Finally, ANT obviates being limited to strict categories of macroand micro-, by accepting that networks may bridge both. This research shows that two rural communities have unde rgone similar growth trajectories, with a historically black community h aving experienced some setbacks in the early 20 th century. However, the results show that the rural African American community was not more subdivided than the neighboring Euro-American community, contrary to initial expectati ons. Additionally, there is a suggestion that communities may move socially importan t buildings such as churches schools to the community center or periphery, dep ending on the intended recipient of the message. The study also docum ents the centralization, concentration, and clustering of the county’s African Amer ican population through time.


1 Chapter 1: Introduction The research presented in this document is a comparat ive study of spatial relationships in the built environment, between a rur al community settled by freed slaves (Bealsville, Florida), the neighboring settleme nt of their former owners (Springhead, Florida), several historically black urban neighborhoods (Tampa, Florida), and a company town that provided onsite re ligious and educational services for its African American employees (Willow, Flo rida). The comparison of a small number of settlements, in a restricted geograph ic area, is only a beginning. However, it is assumed here that “[s]ignifi cant patterns or ideas will not emerge without comparison of differences over time or space (Kowalewski 1997:291). The reason to incorporate a GIS approach i nto such a study is that it allows for data, from both historical and archaeological sources, to be easily integrated into a body of research, and for those data to be examined at multiple scales, allowing the researcher to avoid the “tendency to compress spatial and/or temporal diversity,” while still permitting the “care ful juxtaposition” (Feinman 1997:369,372) of both the historical and archaeological records. Thus, we are not limited to simple descriptions of culture history; w e can also call upon patterns that may not have been observed by those tha t created them, “not by abandoning the comparative method, but by applying i t critically and rigorously at all scales at which the pattern can be recognized” (Marquardt 199 2:126).


2 This research started with an electronic version of the General Land Office (GLO) map, used by the state office’s salespeople (Figur e 1). The GLO was Figure 1. GLO Map of T29S, R22E formed in 1812, and “it supervised the use and disposit ion of the huge federal landholdings until 1946, when it was combined with the Grazing Service to create the Bureau of Land Management (Thompson 2000:289-290 ). Although the GLO map is a two-dimensional representation of land owner ship, the deed recipients


3 had met requirements to improve that land by buildin g dwellings and planting crops. Ultimately, the units of analysis for the rural areas of eastern Hillsborough County, during the earliest period of settlement, are the deeded property owners shown by the sales plat map of the GLO that was used to record legal public property transactions from 1825 until 1907 (Florida D epartment of Environmental Protection, Division of State Lands, Board of Trustees L and Document System 2007). The GLO map, in conjunction with GIS software, offers an opportunity to explore the hypothesis that the original settlers of B ealsville placed the edifices of their built environment purposely to express their str ength, ideationally, as a community based on the explicit virtues of agriculture, religion, and education. This research suggests that the built environment is a f orm of non-verbal communication, and that by analyzing the patterns of cen tral tendency and variation found in measurements of distance, taken from a series of historic maps, between residences and structures that provided services to settlement residents, we will be able to document a means of differentiat ing similar communities that were founded by Baptists from different socio-economic b ackgrounds. We also suggest that by understanding the patterns of use and o wnership of particular properties, whether as rural farmsteads or urban house lots, we can better understand the shape of those settlements in the present Finally, we also attempt to resolve the usefulness of a GIS as a framewo rk for creating a geohistorical, spatially organized, database of multip le documentary forms. As a whole, this paper is a lengthy suggestion of how those d ata might be collected,


4 how they might be analyzed, and how they might provi de a basis for further research that includes descendant communities. To move the study of a local community forward with GIS maps and other records, it is necessary to build a holistic profil e of the people that originally settled that community. Creating a GIS version of the GLO sales plat for Township 29 South, Range 22 East (T29S, R22E), as recko ned by the Public Lands Survey System (PLSS), offered the opportunity t o compare an historically black settlement, dating to the early the Reconstruction Era, with a white “parent” community where the slaves had been held in captive. T he initial allocation of properties provides insight to a particular period of time, so the divisions of properties were evaluated for later periods, whenev er comparable records were available. The rural settlements discussed here were comp ared with similar settlements found through literature review, to conside r how local patterns of property ownership and the use of space may have diff ered from elsewhere in the South. The study also considers several historically black urban neighborhoods and a nearby company town, to better un derstand local patterns in the allocation and use of space where the materia l inertia of the built environment and/or lack of property ownership may hav e constrained the ability of people to physically reconfigure their settlement. However, the ultimate success of GIS-based approaches depe nds on researchers including data that go beyond the individua l project’s research design, to incorporate all the information that has be en compiled for a particular locale into a scalable framework. Researchers must also b e willing to share


5 geospatial data, to begin to confront the “distortion of perspective that comes from tracing phenomena back in time until they disappea r from view over a temporal horizon that lies beyond the interests of the observers or their preferred techniques of observation” (Bailey 2008:25). There ar e, admittedly, issues that could arise from sharing accurate information about the location of archaeological resources indiscriminately, but the increasi ng prevalence of mapping technologies that are accessible to the general public should also facilitate information sharing. For the security of th e archaeological resources, the proper use of the asymmetry of knowledge would all ow for interested stakeholders to share intersite and intrasite geospatial l ocations that allow archaeologists to develop accurate depictions of archaeol ogical contexts, to provide a basis for all stakeholders to assess ongoing her itage management needs. The need for greater time depth in studying the patt erns of history was certainly understood by Marc Bloch when, for example, h e chastised those who claimed to understand “the contemporary economic system on the basis of observations limited to a few decades,” and yet that is still frequently the time scale that is considered in policy-making decisions (1992:42 ). It is a shame that Bloch did not survive World War II because his own insist ence that “a historical phenomenon can never be understood apart from its mom ent in time” had brought him to the conclusion that archaeology had to b e included in any useful study of history, even relatively recent history (1992: 29,44). This was most eloquently stated in The Historian’s Craft (written during the war):


6 Successive technological revolutions have immeasurably wid ened the psychological gap between generations. With some reason perhaps, the man of the age of electricity and of the airplane fe els himself far removed from his ancestors. With less wisdom he has been disposed t o conclude that they have ceased to influence him (Bloch 1992:30). Beyond any value as an archaeological project, or as an ongoing anthropological study of how people engage with their local community histories, one of the most startlingly clear trends in the following research is th at policy is often made without considering any advice at all from historic tren ds, of insufficient time depth or otherwise. The reason that Marc Bloch gave for historians to more closely adhere to methods of archaeologists is one that any archaeologist sh ould quickly recognize: history, as it is recorded, is unreliable enou gh that it must be independently confirmed, especially through the interp retation of the material residue of past actions. The recognition is that the er ror is inherent in the recording process, although errors may also be introduced or compounded by the interpretation process. Worse yet, the error only becomes more entrenched with the passage of time, and with each inveterately recited “fact” called forth from some past educator’s checklist. All of that considers only chance errors in statements of fact, not the errors of omission. Written history provides no defense at all against whol e groups of people being consciously and systematically excluded from history’s records altogether. We know that the preservation of material is neither consistently representative,


7 nor necessarily a complete picture of all past behaviors and full range of attitudes toward those behaviors, even given an ideally represen tative sample of all archaeological deposits. Furthermore, historical records a re replete with examples of marginalized people living on the periphe ry of a neatly stereotyped “normal” society, and those records make clear a consistent pattern of “making do” with materials that were unlikely to be preserved in any primary contexts due to ongoing reuse, with materials that were of substanda rd durability given the expected relative conditions of a specific time and place and with materials that are likely to accumulate more rapidly at certain locales because socio-cultural conditions were imposing boundaries over territories tha t may not be expressed in standard geo-political maps from historic periods. In an article entitled “Saving a Few, Before Losing T hem All: A Strategy for Setting Priorities,” a GIS approach was used to organiz e information about properties from different time periods, to characteri ze the environment in which those properties are located, and in setting preservatio n priorities for a property that is already under management by an agency with som e agenda for interpreting the past for the benefit of the public ( Osman 2002). The use of a GIS is commendable for organizing spatial data at differen t temporal scales, and for combining both environmental and cultural elements int o a single database, to frame interpretive goals. Unfortunately, beyond th e bounds of the properties which have already been designated as eligible for gen eral preservation, such as the National Recreation Area (NRA) discussed in the afor ementioned article, are many properties that are equally deserving of being p reserved. However, many


8 are not being managed at all, either because they ha ve not been identified as having sufficient ecological or historical significance, or because they are not located where regulations mandate the development of a management plan. Even beyond preservation, many properties are not bei ng considered as eligible for conservation of historical resources, becau se there are insufficient tangible remains associated with those properties present ly to engender professional interest, unless an interested group has mobi lized to publicize their objection to a perceived threat. Therein lies the sour ce of the problem; those who were marginalized in the recent past stay marginal ized in the present and into the foreseeable future because the rules regardi ng the identification and designation of historic sites are not met by sites where t here have been undeniable palimpsests created by the very policies of d enial and exclusion that have continued unabated precisely because of the refusa l to acknowledge that there are lessons to be learned from the sites in questio n. One promising approach may be to consider such places as Traditional Cultural Properties (TCPs), since that allows for people to articulate why a place is important to them, cultivates a conservation ethic, and stresses partnerships be tween interested stakeholders, cultural resource managers, and agencies of c hange (King 2003). However, to be truly effective, there needs to be a m ore proactive approach to identifying potential TCPs well before there are pro posed changes to those properties. The use of GIS in predictive models to locate areas w ith a high probability of containing archaeological sites, according to some for mula of what types of


9 environmental features one expects to find at a given site type, has been sufficiently demonstrated. Zehra Osman’s article illustra tes another use for GIS as a planning and management tool for sites that are a lready managed, but perhaps it is also a use that illustrates greater possibil ities for using GIS in cataloging and analyzing both localized individual site s and diachronically varied networks of sites across regions (2002). It should not re quire a last resort prioritization of what can be preserved, and what must be let go, to suggest that there is a value in gathering together cotemporary soci o-spatial information, with historic socio-spatial data reconstructed from material d eposition, and also spatial data drawn from deeper time structures. The u se of GIS should be widespread as a tool for bringing together data from “the study of objects and their [spatial] relationships,” and also from the “in vestigation and interpretation of what may be called the ‘cultural matrix,’” which incl udes both the geographic and the temporal dimension of those features across the env ironment (Taylor 1983:175). Taylor’s conjunctive approach was not a steri le comparison of empirically categorized materials, “but rather [a compa rison of] cultural contexts and/or broad cultural complexes as wholes,” (1983:168-1 69) along with the use of quantitative analysis to reduce warping and errors in data. As a tool for organizing and representing spatial inf ormation in three dimensions, it seems clear that GIS are an excellent too l for evaluating local cultural matrices in fine, diachronic detail. More impo rtantly, they seem ideally suited to assessing culturally appropriate (as opposed to t he manufactured empirical categories of the researcher) boundaries at a nd above the level of


10 regions. Another positive benefit of using a GIS is th at large portions of a site’s history can be established through non-invasive methods, such as magnetometry, ground penetrating radar, electrical resistivity, geore ferenced historical maps, historical aerial photography, and even through surface survey data. That is yet another indicator of a deeper use for GIS, as means fo r bringing the full “culture matrix” to bear on a site’s evaluation for historical si gnificance. Along with the proven record of utility in modeling probable sites, that should also make GIS a useful tool for taking a more cooperative approach to heritage management. It is imperative to identify potentially significant l ocal sites by compiling all available data, then match those sites with extant pub lic and private special interest groups that could be mobilized to advocate fo r specific sites to be brought under some form of heritage management. The increasing availability of internet-based mapping sites (e.g. Open StreetMaps, Go ogle Earth) should provide a means of connecting archaeological professionals with an interested public. That is not to suggest that every location tha t is identified needs to be automatically accepted as an archaeologically significant en tity. A collaborative, computer-based environment does not eliminate any ind ividual’s biases, whether professional researcher or public participant. It does of fer a space for negotiating the identification process, and for assessing the data that are potentially available for inclusion in a particular study. Moreover, the da ta can be separated into thematic layers that can be added, removed, or modifie d to include only what is relevant to a particular line of research. The impor tant point is that the emphasis is on providing a contact point for communicating areas of potential interest.


11 This isn’t simply a call to some sort of “activist” archae ology to right the wrongs of past social injustices by putting them on trial in the present, and it is not being suggested that archaeological interpretations should be fitted to an assumed public good. Although, the purpose of choosing the particular case of African American settlements throughout the American Sou th, from the Reconstruction to the present, is precisely because the econo mic disparity between the South and the other economic regions of t he U.S. still exists, and is mostly due to more than 145 years of policy-makers putti ng political and personal enmities above the well-being of the people living i n their various constituencies. The cost of passively waiting to be asked where the archae ologically important resources are before taking action to officially recognize interpret, and preserve those resources, is not just the loss of those resources befo re they become eligible under some arbitrarily assigned requirement, but in the long-duration and systemic breadth of the negative ramifications of not s ooner recognizing the deleterious social trends. If Marc Bloch had survived Wor ld War II, then perhaps we would have sooner realized that the reaching out t o understand history in its own contexts to, “in one way or another, aid us to liv e better” (1992:9). In other words, we are not stewards of some discrete and “dead” p ast; we are servants to the extant communities in which we live and work. We, as archaeologists and other stakeholders, are still stewards of the material expression of past human behaviors, to whateve r extent they are identifiable, interpretable, and eligible for prese rvation. However, especially with more recent historic sites, a large part that chain of m ateriality includes


12 documentary evidence, which often has similar preservatio n issues and biases. As material culture, many documentary historical records are perhaps not faring so well. In many cases, the historical records are ultima tely being stored in some type of electronic format, essentially in a condition o f “pre-materiality” ready to be brought into the material world again only when the y are printed out by an end user. Indeed, the singular documentary historical recor d that served as the initial focus of this paper entered into the research as an elect ronic document, and throughout the course of working with the GIS software that was used to catalog spatial information represented on a map, several fail ed stages existed briefly only in a non-material state within the software envi ronment (that immateriality precludes those images from ever being made material i n exactly the same way again). O.V. Burton has identified GIS as “a vital area wher e rural historians need to use the computer” to build a diachronic community h istory on the local level, through maps (2002:651). Computer generated maps of fer myriad possibilities for visually representing, three-dimensionally, everyt hing from past built environments to past natural environments. In fact, t hey offer an ability to visually drive home, for the viewer of the resulting image, many things that were not readily conveyed by two-dimensional, paper maps. W hile one of the outcomes of this research is the creation of a database of paper maps converted to electronic GIS layers, there are ramifications to l eaving behind more permanent paper maps. Unfortunately, that may also mean that the easy manipulation of locational data, and the representati on of limited aspects of a


13 geographic area’s physical reality, must always create map s that are intentionally biased by emphasizing where the map creator wants the r eader to focus his or her attention. The listed properties on the east-central margin of the county contain several properties listed in an unincorporated portion of the county that is referred to here as Bealsville. Although the commun ity of Bealsville has never been officially incorporated, it has existed as a histori cally black community since the 1860s. However, the community also has a connectio n to the historically white community of Springhead, just to the north of Bealsville, as many of the former slaves had been owned by the founders of Spring head. In an effort to start building an inventory of culturally and histori cally significant sites, with specific relevance to the African American community both in Hillsborough County and more generally to Florida and beyond to postbellum African American settlement patterns, we must first ascertain how the settlement pattern seen in rural Hillsborough County compares with other p ostbellum settlement patterns. One commonly asserted pattern in postbellum settlement for recently freed people in the South, is that they preferred hi ghly dispersed settlements during the early years of the Reconstruction (especially the late-1860s to late1870s) with a tendency toward settlement nucleation on ly after the turn of the 20 th century (Aiken 1985:387-392). Notable exceptions to the pattern occurred in areas of Gulf Coastal Louisiana where sugarcane agricultu re had dominated (Aiken 1985:387-392). Since the physical structures refle cting the built


14 environment of the original settlement at the locatio n now known as Bealsville no longer exist, the location of the settlement’s origina l structures can best be understood through historic land ownership documents, h istoric maps, and historic aerial photography. In the case of the Bealsv ille settlement, oral history was informally supplied by information booth attendan ts and other attendees of an October 2007 reunion at Bealsville’s Historic Glover School. If the African American settlement pattern at Bealsville reflects the t rends seen throughout the South during the Reconstruction Era, then the settleme nt should be at least as dispersed (if not more dispersed) as the associated Euro-A merican settlement of Springhead. In any case, the data considered here will form the basis of a GIS catalog of the Bealsville site to be used as a prelimin ary guide for “ground truthing” these locations for archaeological significance and also serve as a record of some of the earliest African American settlers in Hillsborough County to be granted a legal deed representing ownership of la nd that had been improved for use as agricultural and domestic residential purposes. Bealsville is also an ideal location for demonstrating what can be accomplished through an equal partnership between a uni versity-centered anthropology department and a knowledgeable descendan t community. Moreover, Bealsville is a consummate example of a commun ity that has independently taken collective control over their heri tage resources. Thus, they are already in a position to be full partners in an o ngoing program of research. In fact, the community has already formed a not-for-prof it group under the name


15 Bealsville Incorporated. Through that organization, they are currently administrating their known resources within the community Ultimately, the questions being addressed here are ab out the spatiotemporal nature of historically black settlements, whethe r in urban or rural contexts. However, the answers are not expected to be si mple, because the socio-cultural norms, active behaviors, and material iner tias all change at different speeds. Moreover, community connections are n ot easily traced solely within settlements, and with greater time depth associat ions become more elusive. It is assumed that settlement co-residents do form a community, based solely on the need to negotiate the use of space. How ever, the definition of community is much broader than a spatial configuration o f the built environment, which should not be prefigured, and people may be mem bers of multiple, overlapping communities. In this particular case, the units of analysis in the rur al areas of eastern Hillsborough County are, at first, defined by the pro perties deeded to settlement founders (which, in some cases, are from the generation after those who initially settled a particular location). The urban settlements being addressed here have been defined on the basis of later historic reports and, in any case, do not reflect the founders of a settlement, but rather the recipient s of properties that had already been divided into small house lots. Despite those apparent incongruities, settlers in both types of settlement shared connections to settlement co-residents, as well as socially important structures such as churches and schools. So, the research presented here must go beyond simply mapping r elationships between


16 structures in the built environment, to determine that non-verbal messages may be encoded in the measurable relationships of those str uctures.


17 Chapter 2: Physical Setting The sites discussed herein are scattered across Hillsborough C ounty, hence most of the county’s physiographic features are in cluded in the study (Figure 2). The rural study area is, mainly, the ea stern 18 Sections of Township 29 South, Range 22 East (T29S, R22E), and both of th e East Hillsborough settlements discussed herein are in the Polk Upland, which is separated, by a scarp, from the Gulf Coastal Lowlands to the west and th e DeSoto Plain to the southwest (White 1970:132-133). The high ground in th e rural study area is around 30.5 meters above mean sea level, and the low ground created by water features is around 21.3 meters above mean sea level. The present-day Bealsville location sits primarily over poorly drained Scranton and Portsmouth sands; however, the earliest town location, just to the south, sits over a large exposure of well drained Norfolk fine sand (Hillsborou gh County 1941:56). Even though the town location did shift slightly northward, later in the history of the community’s development, the entire study area is situat ed over the Bone Valley Member of the Peace River Formation, which underlies most of the Polk Upland (Campbell 1984:1-2). The Bone Valley formation is highly siliclastic, and resi stant to loss to solution in groundwater. This is the reason for the p revalence of stream branching seen throughout the study area (White 1970:1 33). The Bealsville


18 Figure 2. Physiography of Hillsborough County, Flor ida


19 community location corresponds to the southeastern quadra nt of T29S, R22E, and the Springhead community location corresponds to th e northeastern quadrant of T29S, R22E. The entire study area’s main natural community is pine-dominated Scrubby Flatwoods, but there are also S pring Run channels created by springheads originating in both quadrants o f the map discussed in this study. The eastern half of T29S, R22E is geographicall y circumscribed by a natural chain of lakes to the north and west (which actu ally deflected the placement of the Central & Peninsular Railroad to the western half of T29S, R22E); to the east and south of Bealsville, the area i s geographically constrained by the North Prong of the Alafia River (including Ho well’s Creek, Howell’s Branch, and English Branch). Tampa, Florida is, primarily, located in the Gulf C oastal region of peninsular Florida and the Western Valley through wh ich runs the Hillsborough River. The historically black neighborhoods under discussi on here sit astride Township 29 South, Range 18 East, and Township 29 Sou th, Range 19 East, in Sections 12-14, 23-24 of the former, and Sections 7-8, 17-19 of the latter. The neighborhoods listed as West Tampa sit just above the wat erline of the Hillsborough River. The neighborhoods of Dobyville/We st Hyde Park, the Garrison, West Palm Avenue sit about 4.5 meters above m ean sea level. The neighborhoods of the Scrubb and Ybor City sit about 7. 5 meters above mean sea level; and, the neighborhoods of College Hill and Ro bles Pond are just below 15 meters above mean sea level. The neighborhoods are u nderlain by rubble fill,


20 and relict marine terraces; the terracing is more develo ped to the south of the Hillsborough River (White 1970:144). The remnant of the company town of Willow is located on the DeSoto Plain, at Township 30 South, Range 20 East (T30S, R20 E), Section 31, which is underlain by the same Bone Valley Member of the Peace River Formation that rests beneath the Polk Upland (White 1970:133). Only a small portion of the northern reaches of the DeSoto Plain extend into Hil lsborough County. In this area, elevations average to about 23.7 meters above m ean sea level, and the tendency is toward being extremely flat overall (Whit e 1970:140). The channel of the Little Manatee River runs along a northwest-southe ast axis, approximately 200 meters to the east of the town. A small, unnamed linear water feature forms runs along the northeast boundary. The town’s structure s are built on a gentle rise, from northeast to southwest, of 9-14 meters above mean sea level.


21 Chapter 3: Historic Background Prior to the Civil War, most of the American South w as inclined toward the plantation being the economic focus for the wealthy land-owners, and both free, white workers and owned, slaved labor. As a consequence, the South’s major urban centers were located on its coastal periphery, and those cites were focused on exporting the South’s main cash crops of tobacco rice, sugarcane, and cotton (Hilliard 1994:115-117). The resulting distinction between rural settlements being focused on the plantation, and the ur ban settlements being focused on primary trading centers, encouraged the slave holders to increase the profit realized from their slave holdings; and, “especi ally during the last two antebellum decades,” this led to a policy of “hiring ou t” slaves (Goldfield 1991:130). The fine details of “hiring out,” were l eft to the slave-holder, the person renting the slave’s labor, and possibly the slaves themselves; slaves that had been “hired out” often made their own living ar rangement for at least part of the year, even where this was technically against local l aws (Goldfield 1991:137140). The combination of having little in the way of money to rent nicer properties, and the fact that “hired out” slaves weren ’t legally entitled to “living out,” pushed the population to the periphery of the urban centers “outside of areas controlled by fire codes that prohibited the ere ction of flimsy wooden structures (Goldfield 1991:140-141). However, that mov e toward peripheral


22 settlements was itself a change from the pattern that ha d been more typical “during the height of urban slave holding,” before t he practice of “hiring out” became more common (Groves and Muller 1975:173). In S outhern cities that were prominent long before the Civil War, the earli est settlements were more likely to have been integrated, to facilitate “the ef ficacy of slave control” (Groves and Muller 1975:173). On the large plantations that comprised large areas of the inner American South, the antebellum trend had been toward large fields for planting, with nucleated habitation areas; during the postbellum period many former slaves rejected nucleation of their habitation st ructures, and many houses were moved closer to a tenant farmer’s assigned parcel of the now subdivided plantation properties (Hilliard 1994:122-124). Postbellum Settlement Patterns in the American Sout h Following the Civil War, there was an overall shift i n population demographics; generally, people were moving into urba n areas, with a concurrent tendency to move out of the American South. The American South is often defined differently by different researchers; u nless otherwise noted, the region includes Alabama, Arkansas, Florida, Georgia, K entucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, and Virginia. The perceived emigration of African Americans, from the South, varies depending on how one defines the South, and whether one treats the South as a totalizing region, or considers differential rates of migration for internal subregions of the South. In particular, high levels of p ostbellum emigration amongst


23 African Americans, during the 19 th century, only holds true for the “Atlantic seaboard states” (Roback 1984:1189). Moreover, in the two decades following the Civil War there were high levels of immigration, for blacks and whites, to the sout hwest of the cotton growing heartland, especially into Arkansas and Texas. For Florida, immigration of black and white settlers remained high into the 20 th century (Roback 1984:1188). In 1870, 85.3 percent of all African Am ericans resided in the South, by 1910 that number was down to 82.8 percent, by 195 0 only 61.5 percent of African Americans lived in the South, and by 1990 that figure had dropped to 46.2 percent (Shelley and Webster 1998). In part, th e retention of existing African Americans, or the attraction of new immigrants, may be related to “federal land sales and not the breaking up of plantations” (Pa rker 1980:1034). However, Florida may have been attractive to recently freed pe ople due to the generally sparse population, especially outside the main cotton pr oducing counties in the extreme north-center of the state, and also the parity between black and white settlers entering Florida (African Americans were 48.9 p ercent of the population in 1870, and 41 percent in 1910), although that bala nce would shift heavily in the favor of whites during the second half of the 20 th century (Shelley and Webster 1998:168). The earliest shift of populations was not from South t o North or West, but from rural areas to urban areas; by 1870, urban areas “with populations of more than four thousand in 1870 saw their Negro populatio n increase by an average of 80 [percent] over 1860 levels” (Kellogg 1977:312). R elative to the total


24 populations throughout the South, native blacks and w hites moved to the cities at comparable rates. For 1870 (7.7 percent whites/8.8 pe rcent blacks), 1880 (8.3 percent whites/9 percent blacks), 1890 (12 percent whites/1 3.5 percent blacks), 1900 (14.2 percent whites/15.5 percent blacks) and 1910 (19.5 percent whites/19.7 percent blacks) African Americans were moving i nto the cities at slightly higher rates, but by 1920 (25.4 percent white /23.5 percent blacks) the trend changed, with more whites moving into the citie s (Roback 1984:1190). However, in terms of overall population, there were compositional difference between North and South. “[B]efore 1900 blacks rarely formed as much as 5 [percent] of the total population of Northern cities, whereas they comprised more than 40 [percent] of the populations of Southern citie s” (Groves and Muller 1975:176). If Oklahoma is included in the South, and immigrant populations are considered, then the South’s residents were 31.4 percent urban in 1930, 36.7 percent urban in 1940, 44 percent urban in 1950, and 57.7 percent urban in 1960 (Nicholls 1964:27). By 1970, 74 percent of all African Americans in the U.S. lived in urban areas; and, by 1980, that number increased t o 85 percent (Aiken 1985:383; Bryce et al. 1977:158). Furthermore, the trend was for African Americans to move into the older city centers of urban area. During the decade from 1960-1970, “[b]lacks increased as a percent of central city populations from 16.4 [percent] to 20.5 [percent],” and “for central ci ties whose metropolitan areas had a population of one million or more, the percent of blacks increased from 18.8 percent to 25.2 percent (Bryce et al. 1977:158). Overall, there were four main settlement patterns i n the urban South:


25 1. alleys and central courts, 2. highly dispersed residences in predominantly white re sidential areas, 3. urban clusters, and 4. “shantytown settlements near the city’s edge, [that] became widespread during 1850’s” (Kellogg 1977:311). The initial settlement was typically followed by newco mers “crowding into existing black residential areas,” which would eventually become o vercrowded and dilapidated; with the increased stress of high populati on densities and deteriorating material conditions, there was a freque nt pattern of abandonment in contiguous neighborhoods, which created opportunities to expand existing areas (Groves and Muller 1975:178). Thus, the two primary means of increasing residential areas were to acquire recently vacated hou sing along the boundaries between existing neighborhoods, and to build “first-occu pancy housing adjacent to the outer fringes of the core urban clusters;” in the case of new growth, “these communities take the form of sectors bounded by highway s and, more frequently, railroad tracks” (Kellogg 1977:320). In Florida, “[b] arriers to expansion could be natural features such as lakes and swamps or cultural featu res such as canals, railroads, and highways;” but, occasionally, if the growt h of black residential enclaves threatened to encroach on white residential a reas, then physical barriers, like walls, were sometimes erected to separate t he communities, such as seen in Boca Raton (Lee 1992:382). Prior to the tu rn of the 20 th century, the division and sale of house lots was typically left to th e discretion of the property owner; however, from 1910 to 1917 racial zoning was e mployed to create racially


26 homogenous neighborhoods, and until 1948 homeowner asso ciations were allowed to utilize racially exclusive covenants to racia l homogeneity (Gotham 2000; Silver 1997). Thus, the patterns of residential centralization, seen in the present, can often be attributed to “the coalescence of enclaves,” as historic settlements become aggregated through steady accretion (G roves and Muller 1975:190). Historic Context for Using GLO Records in Florida Much of the land that was claimed on the eastern half of T29S, R22E, by people of any race, was purchased under the 1866 South ern Homestead Act, from the Florida Internal Improvement Fund’s 4 mill ion acres (at $1 per acre), or under the 1869 Act to Establish a Uniform System of Com mon Schools and a University (at $1.25 per acre), in several cases (Gates 1 940:326). On a national level, the Southern Homestead Act of 1866, which exclud ed the cash sale system from the public lands of five Southern states (A labama, Arkansas, Florida, Louisiana, and Mississippi) was meant to address several conce rns, including an attempt to prevent the postbellum South’s black populat ion from becoming heavily reduced to tenant farming as had been seen in Kansas and Nebraska, after wealthy land barons had acquired vast territorie s (Gates 1936:670-671). The major proponent behind the passage of the 1866 S outhern Homestead Act was a Congressman from Indiana named George Washington Julian. The Act was quite intentionally meant to prevent the old lan ded aristocracy from buying all the public lands in the South after the Civil War leaving the newly freed slaves with only migration or tenant farming as options (Pop e 1970:201-203). The Act


27 was, ultimately, not very successful at establishing widespr ead African American land ownership due to strict regulations about how much land could be claimed (based on “service” to the Union during the war), and also because being granted the deed to the land required an expensive trip to t he closest GLO regional office, which meant Tallahassee for homesteaders claiming land in Florida (Pope 1970:206-207). Of the estimated 400,000 blacks living in the 5 states af fected by the 1866 Southern Homestead Act, only 4000 blacks had made homeste ad claims by late 1869 (Pope 1970:205). However, just over 75 percent of those 4000 claims were made in Florida (Pope 1970:205). To have been gran ted deed to property, the grantee had to have both made the journey to Tallah assee, and have improved the property to be granted. One reason given for t he success of freedmen in Florida, compared to the other states affected by the 1866 Southern Homestead Act, was that extra relief was provided by the Freedma n’s Bureau starting in October of 1868 (Pope 1970:206). The distribution of relief supplies to frred people that were not land claimants was not necessarily u nique to Florida, but the disproportionately high number of land claims made in the state does seem to support Pope’s assertion. The 1866 Southern Homestead Act was in effect for 10 ye ars; although its socio-political intent was to place emancipated African Am ericans on arable farm land in the South, there were constant problems with d ummy claims registered by proxies operating on behalf of lumber companies an d sales instabilities due to railroads (e.g. the Central & Peninsular Railroad, la ter Seaboard Airlines, had


28 extensive land holdings throughout the eastern half of T29S, R22E) and other land speculators (Gates 1940:327-328). The final versi on of the bill allowed for 80 acres to be claimed for the first two years, and 160 acres thereafter. The purpose for making such limitations was to discourage Nort hern whites and newly arriving immigrants from making claims in the So uth, along with former Confederate whites being entirely restricted from claim ing land from claiming land in the South, was to make it more attractive to t he freed slaves (Pope 1970:203). General Oliver Howard, commissioner of the Freedmen’s Bureau made a report to congress a month prior to the passage of the bill; in his report, Howard explicitly said, “if this goal were to be reali zed freedmen must be furnished with transportation, food, shelter, and farm implements” (Pope 1970:204). That recommendation was not given much co nsideration, as Congress focused on punishing the new representatives fro m the former Confederate sates; however, in the sites under direct stu dy here and the case studies used to support the study, wherever adequate supp lies were afforded the new property claimants, those settlements were far mor e likely to survive into the present. Ultimately, another reason that Florida may have seemed particularly attractive to African American settlers is because, here, “cultivators of 10 acres or more of land,…were made eligible for monthly rations until their first crop was harvested” (Pope 1970:206). Settlement Histories of Bealsville and Springhead The two rural communities under study here, Bealsville and Springhead, share many aspects of historical development, as well as ot her socio-cultural


29 factors, that connect them from the antebellum period o nwards. The original founders of Bealsville were freed slaves, many of whom had been brought to Florida by the families operating plantations and/or cattle ranches around Springhead and other communities in northeastern Hillsb orough County. Initially, the Bealsville community consisted of 19 households. Ho wever, the cook for the Sistrunk family and daughter of Claracy Stevens, Rosa S tephens (or “Rosina” in other documentation), was one of the earliest African A mericans in the area to be granted deed to property in the area in July of 188 1 and should be included among the community founders (Brown and Brown 2003:23 8). By the time the property of the study area was sold of f to private owners, the Central & Peninsular Railroad, or the Plant Inve stment Company in the mid1890s, there were 19 households, headed by African Amer icans, in the Bealsville community. There were also 17 households headed by Eu ro-Americans in the southeastern quadrant of T29S, R22E, who are considered residents of Trapnell. In the northeastern quadrant of T29S, R22E, there we re no African American property owners, only 24 white property owners, 4 of whom are considered part of Trapnell. In fact, the northeastern quadrant of T 29S, R22E is conspicuous for being the only quadrant not to have a school for bla ck students (indeed, no services for African Americans) during the Reconstruction Era, with Turkey Creek covering the northwest corner and Hopewell covering the southwest. In 1870, there were 546 African Americans living wit hin the boundaries of Hillsborough County, and there were only 5345 acres un der cultivation by all residents of Hillsborough County (Landry et al. 2004: 20). A decade later, in 1880,


30 there were 757 operating on 13,383 acres across Hillsboro ugh County (Landry et al. 2004:21-23). Among the majority of land owning whites in rural, agricultural areas cattle ranching was dominant from the end of the Civil War (Landry et al. 2004:20). In 1888, large concentrations of pebble pho sphate were discovered in the soils of the Alafia River drainage basin; the disco very of phosphate spurred development of the Port of Tampa, and fueled the ra mpant land speculation that sought to foreclose on impoverished farmers throughout H illsborough County, especially following the severe frosts of 1895 (Landry e t al. 2004:28). W.E.B. DuBois asserted that the typical sharecropping relationsh ip that dominated African American life, throughout the “Reconstructed So uth,” at best lead to a form of “semi-proletarianism” that approximated wage labor, and implies that many communities failed because a strong, land-owning, black bourgeois class failed to materialize throughout most of the South ( Marler 2004:125-126). Bealsville represents an area where African American own ership of farmland was a factor. The original settlement at Bealsville had been called the Howell’s Creek settlement, followed by Alafia, followed Lil’ Alafia (after a white settlement about 3 miles down the Alafia River got a post office under the name Alafia), and it was not until the mid-1920s that the community was finall y named Bealsville to honor one of the earliest settlers, one who had a policy of buying what land he could afford during the “land grabs” of the late 19 th and early 20 th centuries and repatriating that land back to descendants of the origi nal settlers, sometimes with no charge to the recipient (Brown and Brown 2003:18). Alfred Beal was not the


31 only resident that took responsibility for helping to ke ep the community intact as a viable socio-cultural economic unit as a Hillsborough Co unty farming settlement. Oral history provided by Herman Hargrett, suggests that his father had a similar policy on a slightly smaller scale than Beal (Klugh 2004 :132). By the early 1920s, the community of Bealsville’s residences had shifted just far enough north that it warranted that the church building from the late 19 th century had to be physically moved northward to preserve the building’s place at th e center of the community. The schoolhouse that was initially paired with the chur ch building was not moved; however, the present-day historic property of the Glov er School is on land donated by one of the founding settlers, William Glov er (Brown and Brown 2003:125). The connection between the two communities goes back to the earliest days of antebellum settlement in this region of Hillsbo rough County. Apparently one of the Bealsville founders, Peter Dexter, was respo nsible for saving the life of two children from the Howell family of cattle ranchers living just to the west of the spring feature in the northeastern quadrant of T29S, R22E; it is the same Peter Dexter that allegedly had acquired survey skills during the slavery period to be able to plat the Bealsville properties accurately enou gh to satisfy the GLO (Brown 1997:42; Klugh 2004:109). In any case, both communiti es were united by the same Baptist denomination of Christianity; it has been noted by previous research that there were fewer and/or less severe episode s of racial tension where associated but segregated African American communi ties had organized


32 Baptist rather than African Methodist Episcopal (AME) de nominations early in the development of the community (Klugh 2004:118). Despite connections through history and religious denom ination, there was still some evidence of distrust, regarding the surro unding white settlers, by the African American community, and two founders may have also been instrumental in setting up more remote African American settlements towards the state’s interior, even though the settlements were neve r fully realized, as severe racial tensions never materialized for Bealsville’s earli est residents (Brown 1997:43). The configuration of the Bealsville settlem ent was impacted by severe winter freezes in the mid-1890s; some residents had pla nted orange groves, which ended up being hit the hardest by freezes (Brown 2000:167). Furthermore, the discovery of phosphate in this region of the county also had land speculators in the area looking to buy property at bargain prices. The end result was that modern Bealsville is slightly farther north than the o riginal community (although the territories overlap) (Klugh 2004:135). Both the South Carolina settlement of Promiseland and the Florida settlement of Zion listed sugarcane farming and cane syrup production as major economic contributors to pro perty owning agriculturalists (Bethel 1981:70; Brown 1994:176). Th e Promiseland settlement is even geographically circumscribed by the same Seaboard Airline Railroad as ran to the along a north-south axis on the western side of the lakes that bisect T29S, R22E (Bethel 1981:71). Hillsborough County seems to have had numerous integ rated spaces for religious practice and, even theatre attendance, datin g back as far as 1838


33 (Brady and Brown 1997:15). However, the Springhead and Bealsville were well separated in regard to the community services that were available to their respective populations. Both communities developed a Ba ptist Church in the earliest phase of community growth, but the major non -religious community services were a gristmill (later, it included a sawmill) run by the English family of Springhead, whereas the for the Bealsvile community it was the Cane Press that was brought down from Wakulla County in the early 189 0s, by O.V. Hargrett (Hillsborough County 1998:37,98; Klugh 2004:132). A frican American residents of Hillsborough County even had some representation i n local politics, with both Mills Holloman Jr. and his son Adam Holloman serving as C ounty Commissioners within the first 20 years after the Civil War ended (Brown 1998:97). The elder Holloman had his term suspended for a brief period, and there is some evidence that his reported illiteracy wa s a problem for white constituents (Howe 1997:84). Another of the other su btle connections between the two communities, is that aside from those born here in Florida, both the white and black settlers were predominantly from coastal Georg ia and South Carolina before settling T29S, R22E; one of the few exceptions was Mills Holloman Jr., who originally came from Virginia (Brown 1998:97; Br own and Brown 2003:3,17,67,89,119,163,226,237; United States Censu s Bureau 1958). Another building at the core of the Bealsville built environment, was the log cabin schoolhouse known at first as the Antioch School, and later as the Jameson School (Hillsborough County 1998:40; Klugh 200 4:141). School was originally held within the structure of the Antioch Ba ptist Church; however, there


34 are some discrepancies regarding the schoolhouse’s constructio n date. One recent report suggests that the original Antioch School was built in 1873, and that the building was moved from its initial position to it s later position on Sam Hicks Road; however, Elgin Klugh’s doctoral dissertation resea rch seems to indicate that a new Antioch school was built around 1891, and o nly after a later room addition was that building called the Jameson School. T here have also been orally communicated reports that the log cabin schoolhou se burned down in 1899, but that date is not otherwise indicated by any evidence ; however, this could be explored archaeologically by relocating the original and second positions of the Antioch Baptist Church and the Antioch School. Destructio n by fire would provide a better explanation for the schools name chang e, which can otherwise only be weakly explained by the addition of a second r oom, in the 1920s; establishing the archaeological integrity of any deposit s associated with the original (group ritual/group service) edifices will b e the first step in modeling the spatial relationships of the other structures distribut ed throughout the built environment and landscape of Bealsville. Comparing Local Rural Sites with Other Areas of the American South One of the enclave communities in the literature tha t is comparable to Bealsville is a community that developed in the Piedmo nt region of South Carolina during the Reconstruction Era, was the communit y known as Promiseland. The location of Promiseland, on 2742 acre s, was originally part of a sales scheme geared to African Americans living in Sou th Carolina; the property was a private holding that was ultimately sol d off in small farm lots


35 following the death of the original white plantatio n owner. Unlike Bealsville, where the land was acquired from government holdings through the Southern Homestead Act of 1866, the land at Promiseland was sold by private agent for a down payment and 3 annual follow-up payments (Bethe l 1981:27). The Promiseland settlement did have a high density of occup ation; in 1897, there were 289 households living on 2742 acres, with a mean p roperty size of 34.8 acres (Bethel 1981:98-99). Promiseland also offered so me comparisons in regard to the number of female-headed households, wit h the South Carolina community reported having around 17 percent of all ho useholds being headed up by females (Bethel 1981:122). Of the 39 original, deeded property owners on the eastern half of T29S, R22E, that were assigned to the communities studied here, 4 were female-headed households; the small sample siz e makes statistical comparisons difficult, but it is notable that similar fre quencies of female-headed existed for both racial classifications (i.e. white and bl ack), with 2 out of 20 white, and 2 out 19 black households having female heads (Uni ted States Census Bureau 1958, 1965, 1978). The number of female-hea ded households is also comparable to other Baptist settlements in Florida; fo r example, the Marion County “Afro-Baptist” settlement of Zion listed 2 fem ale-headed households among their earliest homesteaders (Brown 1994:176). The development trajectories of the Florida and South Carolina communities ended up diverging radically during the la te decades of; Promiseland’s population lost control of enough farmlan d that the entire community eventually shifted to wage labor outside the community, whereas the


36 Florida communities of Bealsville and Zion continued to exist as subsistence truck vegetable farmers, rather than in purely commercia l crop like cotton that is essentially what killed Promiseland as an agricultural com munity (Bethel 1981:173-175). It wasn’t geographic territory circumscri ption that killed Promiseland, because it was situated in the same pine d ominated upland scrub that was endemic to both Florida settlements. There w as roughly 3 times the acreage, around the turn of the 20 th century, at Promiseland (approximately 2700) for 289 households, as the approximately 900 acr es at Bealsville for 19 households, compared to Zion’s approximately 300 resid ents spread out over 1700 acres (Brown 1994:176; Hillsborough County 1941:5 6). It does seem that what sustained the Florida communities was the reliance on pure subsistence level truck vegetable farming that helped to sustain t hem as agricultural communities; however, no black bourgeois emerged among t he Afro-Baptists of the Zion settlement, and it ultimately did not survi ve into the late 20 th century, whereas Bealsville is still a tightly knit community. In a 1985 article, Charles Aiken addressed the question of postbellum settlement throughout the South, from the period of Reconstruction and into the 20 th century; the most important assertion made by Aiken is that freed blacks immediately began to reject the nucleated settlement p attern that had been imposed on them when they lived in the slave pens nea r the “big house” under the old Plantation System (1985:389). Aiken also sugg ests that the one place where the old nucleated settlement pattern did continu e was the sugarcane plantation of Louisiana (1985:367). The Bealsville a rea did have a good deal of


37 sugarcane production, as suggested by a quote from Step hen Harvell in Slavery in Florida: Territorial Days to Emancipation “that just about everybody in the region ’raised a heap of [sugar]cane and boiled it in vats and kettles at different places’” (Rivers 2000:74). Canter Brown also noted th e impacts of sugarcane for local freedmen agriculturalists, along with potatoes an d corn, in advertisements from a small local newspaper called The Newnan Herald that “laborers receive from $25 to $30 per month, or when hired by the day $1.50 a day” (1997:47). For Hillsborough County, specifically, the 1880 census r eports that 556 acres were used for cotton, and 238 acres used for sugarcane p roduction; in comparison, 4968 acres were used to grow corn (Smith 188 4:184). Comparable settlement patterns also exist elsewhere wi thin the state of Florida; for example, a mosaic of small African America n settlements along the Atlantic Coast of Dade, Broward, and Palm Beach counti es. The black settlements of Florida’s Atlantic Coast were more likely to be territorially circumscribed by white neighborhoods to industrial/tran sportation boundaries such as road systems or railroad tracks, or geographically ci rcumscribed by scrub or other undesirable land types (Lee 1992:376-37 7). Most strikingly, Lee notes that Florida’s black population on the Atlantic m argin had two-thirds of their population (about 450,000 people) confined to 97 tra cts; whereas, white populations along the Atlantic margin of south Florid a, spread two-thirds (about 2.1 million people) of their population spread across 2 34 tracts (1992:378).


38 Settlement History of Tampa’s Historically Black Ne ighborhoods The original town of Tampa grew up around the milit ary Fort Brooke, and was far from the representation of an urban built en vironment that it has become; the original fort was constructed in 1824 (Brown 1999:1 4-17). The fort itself is often remembered as the point of origin for Major D ade, whose ill fated expedition to Fort King touched off the Second Semi nole War on December 28, 1835 (Brown 1999:50). The fort would also serve as a temporary interment facility for Indians awaiting transport to Indian Ter ritory in what is now Oklahoma (Brown 1999:48). By the end of the 1830s, several l ots, including 3 on Tampa Street near Whiting Street, had been sold to privat e citizens, and the stage was set for a burgeoning civilian population (Pizzo 1969: 7). The town of Tampa was finally incorporated in 1855, only 10 years after Flo rida was established as a state (Dunn 1977:16). Tampa’s growth progressed slowly through the next 30 years; it was hampered by an outbreak of Yellow Fever in 1858 that reduced the town’s population by one quarter, and by the decommissi oning of Fort Brooke in 1859, though some troops remained stationed there unt il 1882 (Panamerican Consultants 2001:4; Pizzo 1969:81). The 1880 census records that Tampa’s population was 720 people. Only 5 years later, after Vicente Martinez Ybor agreed to move his cigar manufacturing industry to Tampa, the population trip led to almost 2,500 (Dunn 1977:19,21). One traveler described the unpainted ho uses and dirt roads of Tampa. She called particular attention to several do minant features of the town, including 2 saloons and 2 lumber mills (Hewitt 2001:23 ). It was in the vicinity of


39 one of those lumber mills that one of the Tampa area s earliest recognized neighborhoods, the African American enclave known as the Scrub, sprang to life; this neighborhood, “bounded by Scott [Street] on the north, Cass [Street] on the south, Central Avenue on the west, and Nebraska [Avenue ] on the east” housed the highest “concentration” of blacks in the Tampa area, and was described as ‘impenetrable and serv[ing] to remind one of a walle d city’ (Howe 1999:5; Panamerican Consultants 2001:4). The neighborhood is also mentioned during an outbreak of Yellow Fever in 1887, when “the ‘Scrub ’ and Ybor City were brought under the supervision of the local board of he alth’s inspections” (Barker 1984:48). The household organization of this period tended towards inclusion of the extended family, and perhaps boarders as well; in 1880, about half of Tampa’s 178 African American run households were “maleheaded nuclear families” (Howe 1999:6-7). In about 64 percent of th ose 178 households, a male was the sole money earner of the house; however, the 1 880 census did list, for the first time in Tampa, the profession of laundress (Ho we 1999:11-12). However, there had certainly been laundresses working in the area for quite some time; shortly after emancipation a woman named D orcas Bryant had “homesteaded a 60 acre tract” here, and she supported herself by taking in laundry (Hewitt 2001:26). By 1900, there were 4,382 African Americans living i n Tampa, comprising almost 30 percent of the total population. Over the next 30 years, a disparity in population growth would lower this to closer to 20 pe rcent of the total population (Howard and Howard 1994:2). The African American comm unity thrived in some


40 areas, as evidenced by the numerous black owned businesses that grew up on Central Avenue. The strength of this neighborhood w as in its cohesiveness, and ability to mobilize as a community; much of daily lif e was organized around family, school, and church, but there were also a number of mor e civic-minded enterprises and service organizations. Among those organ izations, the Tampa Urban League, founded in the 1920s, would be instrume ntal in the community; also important were the Clara Frye Hospital for blacks, and a weekly periodical called the Tampa Bulletin founded in 1910 and 1915 respectively. In 1880, county census records indicate that there were o nly “two black teachers” (Howe 1999:23) living in Hillsborough County and they were not listed on any official county payrolls. However, due to a dr ive within the community, sufficient money was raised to open Tampa’s Harlem Acad emy in 1889 (Howe 1999:23). The first few decades of the 20 th century saw the number of schools grow to “eight public and nine private institutions b y the mid-twenties” (Howard and Howard 1994: 6). Churches also flourished over thi s time period; “the city’s four oldest black churches” all dated to the second half of the 19 th century, but by 1926, there would be 42 black churches in Tampa (Howard and Howard 1994:6). The stability of the family unit among Tampa’s African Americans “kept alive black culture and undoubtedly brought the city’s A frican American community needed psychological and economic support” (Ho ward and Howard 1994:3). The census of 1900 records that 75 percent o f “black families with children were two-parent households;” additionally, th e divorce rates also seem to have been low—“in 1930, for instance, there were 2, 010 married women


41 between 25 and 34 years of age compared to only 103 divorced females in the same age range” (Howard and Howard 1994:3-4). Altho ugh the strength of the family was undeniable, few of these families could aff ord to own their homes, and “by the 1920s over 75 percent of the city’s African Amer icans resided in rental housing units located in all of Tampa’s black neighborhoo ds” (Howard and Howard 1994:4). A growing and prosperous Tampa was at tracting an influx of predominantly unskilled workers, around 40 percent of wh om were coming from outside Florida (Howard and Howard 1994:2). These ne wcomers inevitably found themselves crowded into neighborhoods like “The Scr ub [which] stood out as a blighted collection of cheap rental units” (Howard and Howard 1994:3). The cost of these rental units were disproportionately highe r (averaging $5.38 per week) than many other Southern urban centers, such as New Orleans ($4.52 per week), Memphis ($3.73 per week), Louisville ($4.54 per week), Charleston ($3.11 per week), and Richmond ($4.46 per week) (Raper et al. 1927:18). However, these prices were somewhat lower than those of Northern industrialized urban centers, such as New York-Harlem ( $7.16 per week), Dayton ($6.00), Philadelphia ($7.95 per week), and Indianap olis ($5.48 per week), although wages may have been correspondingly higher t here as well (Raper et al. 1927:18). The residents of these neighborhoods could look forward to the assistance of their own community, if not from offices of the lar ger city, through the many service organizations initiated by and maintained by T ampa’s African American citizens. Tampa’s African American community has a strong tradition of female


42 activists such as Dorcas Bryant and Blanche Armwood, both of whom were active both in their neighborhoods and in greater Tam pa itself. A large number of women’s organizations and “sororities rendered importa nt civic services that include charity work, social improvement, building hom es for orphans and single working women, and encouraging” appreciation of the fi ne arts (Howard and Howard 1994:4). With strong community support, areas l ike Central Avenue began to prosper. In 1927, the Urban League commissioned a study of Africa n American life in Tampa, which remains the authoritative source for in formation about Tampa’s African American communities in the 1920s. The study was nominally conducted under the supervision of Arthur Raper, but the researc h was in actuality conducted by, and the report written by Dr. Benjamin E. Mays with the assistance of his wife Sadie. Around the time of Mays’ research, “the black business community consisted of some 185 various African American commercial establishments that employed about 400 men and women” (Howard and Howard 1994:8). Some of the notable businesses include the C entral Life Insurance Company, which eventually became “one of the state’s la rgest black enterprises,” and the previously mentioned Tampa Bulletin which was not just “black-owned and operated, [but also] employed only black typesetter s and linotype operators” (Howard and Howard 1994:8). The Tampa Urban League provided a number of services to Tampa’s African American community including “ two day-care nurseries for working mothers, alternative home placemen ts for juvenile


43 delinquents, family case work, the organization of club s and recreation for black youth, and even employment placement” (Howard and H oward 1994:9). The 1927 study examined every aspect of African America n life in Tampa, including education, social services, religion, housing, e mployment, and population statistics; officially the report was titled “ A Study of Negro Life in Tampa: made at the request of the Tampa Welfare Leag ue, the Tampa Urban League, and the Tampa Young Men’s Christian Association ,” however it has become colloquially known as the “Raper Report.” The report included, under the description “Tampa Proper,” both the upscale Centr al Avenue and the impoverished neighborhood referred to as “the ‘Scrubbs; ’” however, the report also stated explicitly that the neighborhoods were “a u nit only in that [they occupy] contiguous territory” (Raper et al. 1927:5). Th is area was home to more than a third of Tampa’s 23, 323 African Americans (Ta mpa’s total population at the time was around 150,000), with a population of 8 ,362 people; the majority of structures in the neighborhood were white-owned rental properties consisting mostly of “1 story frame buildings,” with the “’Shotg un’ type [being] the most prevalent” (Raper et al. 1927:5,11). The study eval uated the living conditions of 326 of these rental houses, reporting that 146 had “B ad” interior conditions, and another 127 were listed as only “Fair;” additionally 259 of these structures had no bathing facilities, and only 114 had an indoor toilet (Raper et al. 1927:11). Outdoor toilets serving multiple families accounted for another 136 properties, leaving 76 with only privies to serve one or more fam ilies.


44 Additionally, only 144 of the houses had an indoor so urce of water; the remaining households were served by either an outdoor spigot (85), or a well (97) for their freshwater needs (Raper et al. 1927:1 1). The City of Tampa claimed that these neighborhoods were served with refuse removal several times per week, everyday in the case of some neighborhoods. However, 44 families reported receiving no refuse removal service, and were likely to have disposed of their own refuse by either burning or buying it in t he backyard (Raper et al. 1927:11). Each family averaged about 4.5 people liv ing in an average of 4 rooms, and roughly half of those families had children under the age of 15; many (31 percent) of those families subsidized the cost of ren t by taking in “lodgers” (Raper et al. 1927:11). In all, Dr. Mays named eigh t separate African American enclaves: The Scrubb, Ybor City, Garrison, West Hyde Par k (Dobyville), West Palm Avenue, College Hill, West Tampa, and Robles Pon d (Howard and Howard 1994:3) (Figure3). Quality of life was indeed problematic for the Africa n Americans in Tampa; there was insufficient hospital space available for the size of the community; during the 1920s, the population was dwindling by shee r attrition, with few births than deaths (Raper et al. 1927:23). Furthermore, d ue to the lack of hospital space expectant mothers had to rely on the skill of unde r-trained midwives to assist in childbirth. Of stillbirths recorded in 1926, ab out 47 percent were attributed to African American mothers; with the black c ommunity accounting for only 20 percent of the total population, there woul d appear to have been a definite disparity (Raper et al. 1927:22). Addition ally, “50 [percent] of Negro


45 Figure 3. Tampa’s Historically Black Neighborhoods


46 deaths were individuals aged15-44,” of the 534 death s (in 1926) in this age group, 42 were directly caused by homicides (Raper et al. 1927: 23). Infants suffered from the squalid living conditions as well; “81 [percent ] of deaths caused by diarrhea and enteritis occurred between birth and 1 ye ar of age” (Raper et al. 1927:26). There was little available to Tampa’s African Americans for recreation; blacks were barred from enjoying the city’s parks “except i n the capacity of servants” (Raper et al. 1927:27). There were 2 theat res on Central Avenue, but both had “extremely bad ventilation and sanitation; less wholesome entertainment was available in places like the Lafayett e Dancing Academy, and the various pool halls that could be found at the back of some of the neighborhood’s barber shops (Raper et al. 1927:27). Dr. Mays summed the situation up quite succinctly when he wrote: For a colored population of 23,000, Tampa provides a Branch Library and a salaried playground supervisor. The City of Tampa p rovides no public park for Negroes: it provides no playgrounds, except une quipped school grounds: it provides no public pool or beach. The priva te recreation and amusements are of such a nature that the Negro publi c receives no benefit therefrom. The Commercial recreation and amu sement is of such a nature and so poorly supervised, that it perchance i s more harmful than beneficial” (Raper et al. 1927:32).


47 Despite being virtually ignored by the City of Tampa these neighborhoods did thrive in some cases; for the most part there was work av ailable, and there were strong community ties to the churches and schools. Large portions of Tampa’s African Americans were emplo yed in unskilled trades, such as laborer, janitor, maid, or bellboy; ho wever, there were also professionals such as physicians, dentists, pharmacists, undert akers, and at least one lawyer. There were also a number of African Americans employed in skilled trades; conditions varied for skilled workers, some, like the “brick layers” were joined with fellow white workers in a union, and received the same pay for the same work (Raper et al. 1927:42). Other skilled w orkers, such as the carpenters and painters, were not allowed admittance to the unions of the white workers; however, there were also non-union jobs where black workers could expect to be well, if not equally paid, including the cigar making operations in Ybor City (Raper et al. 1927:42-43). Additionally, there was the building that housed the Longshoreman’s Local 1402 whose one time le ader, Perry Harvey Sr., gave his name to the park where the current study is located (Howard and Howard 1994:9). Tampa’s utility departments did emp loy black many black workers, but in many cases, it was their policy to “lay off” black workers when whites were in need of employment (Raper et al. 1927 :46). At the time of the Tampa Urban League study conducte d by Dr. Mays, there we “3322 colored children enrolled in the publi c schools,” and another 610 youths enrolled in various private institutions (Raper et al. 1927:53). However, this meant that about a quarter of the school-aged chil dren were not enrolled in


48 any school; furthermore, on any given school day, fully 20 percent of students did not attend classes (Raper et al. 1927:53). The schools th at catered to African American students, in most case, lacked sufficient size or ot her accommodations for the size of their student body; there was also preci ous little space dedicated to recreation, and only the Harlem Academy had any t ype of playground equipment (Raper et al. 1927:55). A 1925 study, cond ucted by Columbia University, determined that almost 85 percent of Tampa ’s African American students were “over age” for their grade level; Mays a ttributed this fact, in part, to irregular attendance and students who got a “late start ” (Raper et al. 1927:55). The schoolteachers responsible for were poorly compensated for their efforts, with salary ranges from $60 to $100 for female teacher s and $65 to $167 for male teachers; out of those salaries, teachers were exp ected to keep themselves attired with “good clothes,” to attend summer classes for self-improvement, and often felt compelled to provide supplies for their stu dents (Raper et al. 1927:55). The church also played a large part in the life of Ta mpa’s African American citizens; even during the late 19 th century, Tampa’s churches were “a major source of news and inspiration, [they] served as a kind of town hall with the minister as publicist and exhorter (Howard and Howard 1 994:5). Mays asserted that, “aside from the home, the church is the most impor tant factor in the life of the Negro” (Raper et al. 1927:48). The churches took on a prominent role in uniting the communities, and a number of ministerial a lliances provided a voice for African Americans in the political sphere of greater Tampa (Howard and Howard 1994: 5-6). Among the most well attended we re Baptist and Methodist


49 churches, as well as “small storefront-like churches that spra ng up in those years;” however, “lower income blacks” were more likely t o become congregants in the Catholic and Episcopalian churches than middle-cla ss or affluent African Americans (Howard and Howard 1994:6). The churches also served as focal points for community action and social interaction; add itionally, the churches were supporters of education, directly sponsoring some pr ivate schools (Howard and Howard 1994:6). From the 1940s to the 1970s, Tampa aggressively pursued expansion, and various urban renewal projects; however, going back to the 1930s “the Scrub” neighborhood was targeted for clearance (Panamer ican Consultants 2001:6). The city annexed a significant portion of th e suburbs surrounding old Tampa in 1953, and began acting on a 1952 ruling in the state legislature that allowed the city to pursue urban renewal projects (Ker stein 1998:77-78). Many of Tampa’s impoverished neighborhoods were targeted fo r demolition; moreover, private housing, rather than public housing, was expecte d to cover almost all the residents that were displaced by these projects (Kerstein 1998:79). A group of Tampa’s prominent African American businessmen, including Perry Harvey Sr., managed to bid successfully for a contract to build housi ng; their organization, Tampa Park Apartments Inc., “developed about 370 subsid ized rental units” (Kerstein 1998:80). “The Scrub” (reported by Mays as “ Scrubb,” which is preferred here for clarity in mapping) neighborhood itself was partially cleared in 1954; the 1968 “construction of the Jefferson Avenue a pproach to I-275,” and other projects in 1970 and 1972 removed the last vestig es of the once


50 prosperous Central Avenue, and the dilapidated structur es of “the Scrub” (Panamerican Consultants 2001:6). Settlement History of Willow Company Town The final site type to be discovered here is a company lumber town that included both white and black employees. The town of W illow, in south central Hillsborough County, housed the employees of McGowin-Fo shee Lumber Company beginning in 1924, although the Foshees sold their share of the operation to the Robbins family in 1926 (Duncan n.d.: 1-2). The mill ceased production from 1929 to 1931, but remained in operat ion until 1936. The town site included just over 80 hectares, with 75-80 residen ces, a doctor’s office, a commissary, a boarding house for single employees, and th e buildings of the milling operation (Duncan n.d.:1). The residential areas were segregated, with the white employees living closer to the main road, and the black employees living closer to the creek that delineated the northeastern boundary of the town. T he “black quarters” also included a church, and a schoolhouse (Duncan n.d.:43). T he white employees attended churches and schools in the nearby community of W imauma (Duncan n.d.:43). The commissary served as a social center, with both white and black employees congregating there outside of work hours (Dunc an n.d.:44). There was, reportedly, a juke joint somewhere between the bl ack residences and the creek (Duncan n.d.:42). The owners maintained residence s in the white section of the town and a small garden that provided fresh p roduce to sell at the commissary.


51 The town received electricity from a generator on the grounds, and it provided power from 5 a.m. to 9 p.m (Duncan n.d.:43) The residences were provided with outside water spigots, unless individual o wners opted to run pipes into their own home (Duncan n.d.:37). The three hou ses occupied by the owners and their kin all had indoor plumbing. The houses we re reputedly of good quality, and were likely sold off to local buyers after the mil l ceased operating (Duncan n.d.:36). The majority of the laborers were paid daily, in tim e checks, which could be exchanged for goods from the commissary or babbitt-ty pe company scrip, which are a “token” used in lieu of legal tender (Dro bney 1996:138). The use of babbitt scrip was typical for company towns throughout Fl orida, but “[p]ayment of wages exclusively in the form of scrip was not common” (Dr obney 1996:138-139). The employees of the commissary, and those who worked in the office, were paid on a monthly basis. All employees had the option to exchange their company scrip for U.S. currency, during one hour, one da y per month (Duncan n.d.:40). In fact, companies were legally obligated “ [u]nder Chapter 6914 of the Acts of the Florida Legislature of 1915” to exchange scri p or merchandise checks for U.S. currency, starting 90 days after issuance (Drobn ey 1996:140). The wages for laborers “were from eight to ten cents an hou r” (Duncan n.d.:40). The use of company scrip is typically thought of as a ty pe of corporate paternalism that serves to reduce overhead by overchargi ng for merchandise that might be available more cheaply to consumers in a competitive market. According to a 1935 study by the U.S. Department of La bor, “it was found that


52 the aggregate average prices charged by company stores we re from 2.1 percent to 10.4 percent higher than those charged by independ ent stores” (Drobney 1996:137). The same study indicated that, “by 1933, o ver 40 [percent] of the annual payroll of Florida lumber companies was recove red through company stores” (Drobney 1996:140). Another means of recoverin g payroll expenditures was by charging a mandatory flat fee for medical care ( Drobney 1996:136). Aside from controlling access to goods and medical care, the presence of company built schools and churches is generally regarded as an element of corporate paternalism, which served to further encourage the immobilization of the labor force (Drobney 1996:129). At Willow, this may have been true for the black labor force, but white workers availed themselves o f the services in nearby communities (Figure 4). The company commissary was a comm on focal point in Florida company towns, and Willow’s appears to have bee n a shared social center for both the white and black workers (Drobney 19 96:136). Other factors that seem to indicate resistance to corporate paternal ism at Willow were the presence of a “bootlegger” that accepted company scrip a s payment, an onsite juke joint, and the availability of prostitutes for th e unmarried employees (Duncan n.d.:42). These locations represent three distinctly different si te types that were available to Hillsborough County’s early African Amer ican settlers. Urban Tampa offered opportunities for property ownership in the e arliest stages of growth, and some flexibility to reconfigure the built environment However, with the growth of Tampa, the built environment became more permanently fixed. The rural,


53 Figure 4. Planview of Willow, Florida


54 agricultural land of eastern Hillsborough County also offered opportunities for property ownership, and flexibility to reconfigure th e built environment as the community’s boundaries shifted. The company town of Wi llow reflects a built environment that was dictated by the owners of a busine ss. Taken together, these sites offer an opportunity to understand how Hill sborough County’s African American residents related to their built environment especially the structures that house socially important services—churches and schools.


55 Chapter 4: Research Background In Rethinking Archaeology K.C. Chang’s simple explanation of theory, as “[providing] us with a basic framework for operating ou r methodology and for determining the actual methods and techniques of gathe ring, ordering, and interpreting data,” should have made it clear that w e are ineluctably bound to be guided by theory in our various research domains (1967: 128). However, more than four decades later, the discipline still seems mir ed in an ongoing discourse about the applicability of theory, generally, to the practice of archaeology, or which theories might provide the most epistemologically and ontologically satisfying explanations, for continued archaeological r esearch (Johnson 2010:217-220; Klejn 2001:1-4). We will ultimately let our particular research endeavors define their own scales of analysis, but there does need to be some means of discussing our research in a non-idiosyncratic way. To that end, this approach advocates using the “scales of behavioral analysi s” (i.e. interaction, activity, and systemic) recommended by the behavioral ar chaeologists, LaMotta and Schiffer (2001:18-19). This research utilizes an Actor-Network-Theory (ANT) ap proach, which is not a theory of the social, but rather a theory about researcher methodology. It has been most widely criticized for conceding the possibil ity of agency to nonhuman actants, but that should not be that alarming fo r archaeologists, after all


56 for us the past human actors are gone and we are left w ith only the material mediators that connected them in their engagement with the living world. What should be the most disconcerting aspect of an ANT inspired research program is its espoused relativism; however, the distinction that mu st be made is that the type of relativism being advocated is not the researcher ’s reflexive, soulsearching moral relativism, but an attempt to keep rese arch constrained by empirically observable associations between agencies in a h eterogeneous network. To quote Gilles Deleuze, ‘Relativism is not t he relativity of truth but the truth of the relation’ (as quoted in Latour 2005:95 ). The primary purpose of this section is to confront scale, and to discuss some approaches t o working with multiple scales of analysis, including both temporal scales and spatial scales. Our primary interest is in establishing how the materia l world impacts the ways people interact and communicate, but that does not mean that human impacts on the material world should no longer remain a viable part of archaeological inquiry (i.e. they are not mutually exclusive). In regard to scale, generally, the ANT position is tha t “[s]cale is the actor’s own achievement” (Latour 2005:185). The archaeologica l ramifications of letting actors define their own scale should be immediately seen. We typically define a scale of analysis in advance, and this can obfuscate pattern s that are not visible within the scales we have pre-selected. Furthermore, th ere is the issue of forcing evidence into a scale that seems well established; a scal e (whether temporal or spatial) which is based on physical evidence of relationshi ps between people and


57 a specific institution may not be coterminous with the b oundary of the same people’s relationship with a different institution. As we shift between households (possibly including multipl e structures), settlements, regions, and supra-regions, we must be prepa red to accept that associations with different institutions are not likely t o create well defined composite groups. The obvious answer is to use scales that a re generalized to be flexible in their use, such as the aforementioned scales of behavioral analysis at the level of individual human-material interactio n, at the level of building interactions into the behavior chains of activities, and at the level of reassembling behavior chains into systems of behavior. However, in a n ANT approach, microscales cannot simply be subsumed into macro-scales, although both share the familiar network shape comprised of nodes and edges. Ta ken to an extreme, any individual person or object would be a node amidst an array of edges connecting it to a larger scale. With the line between what constitutes a microscale or a macro-scale growing ever fuzzier, it might be more useful to think of the micro in terms of how it attempts to “disconnect” it self from the macro, and to think of the macro in terms of how it attempts to keep t he micro “attached” to itself (Latour 2005:180). Since we cannot guarantee t hat complete past networks will be made visible by any single program of research, we may benefit from considering the “ effective scale at which pattern may be comprehended or meaning attributed” (Marquardt 1992:107). Another proposition of ANT that bears on the determi nation of scale, is that actors should be allowed to define, for themselve s, the existence of groups


58 both through declarations of membership in such groups, and through delineation of “anti-groups” that constitute what is outside the me mber group (Latour 2005:32). For example, the rural settlements under discussion here could have been defined by quadrants of a particular PLSS townshi p and range, with any settler within a quadrant (possibly based on an assumed propinquity of structures expressed in a specified spatial distance and temporal concu rrence) lumped in as part of a settlement. To some extent that may have been sufficient to define a spatially related community; but, it would have deni ed how past actors, from the accounts of actual settlers to outsider accounts (such as newspap er articles from the period, or family histories committed to writing a fter the period in question) culled from historical documents, defined their own mem bership in a particular community. The earliest attempts at defining the foun ding residents of these rural communities was based on such a scheme, which created a basis for comparative regularity; however, following more resea rch, historical resources showed that some of the originally included settlers de clared themselves (or contemporaneous sources declared them) members of other se ttlement communities (Hillsborough County 1998; Robinson 1928). So, using a “ cartographic scale” based on the way GLO maps were officially recorded by PLSS units, or a “ geographical scale” based on an “overarching conceptual abstraction lain over these scale d landscapes,” such as the footprint of a house compared to the total catchmen t area of the North Prong of the Alafia River, could lead to manufactured settle ment boundaries that do not reflect the way the space was viewed by settlers (Harri s 2006:41). Analyses may


59 be performed at those scales to answer some research quest ions, but our “ methodological scale” should not be a single, universal scale and it sho uld allow for past actors’ description of their own relationship to metric space (Harris 2006:41). Thus, when conducting excavations it makes sense to use a standard unit, such as the square meter, to assess the intensity of testing as a ratio of the excavated units (where the grain would be the square m eter) to the arbitrary boundaries of the tested area (where the total squared meters would be the extent) (Burger and Todd 2006:238-239). However, i n a GIS analysis based on historic documents, aerial photographs, and human recol lections, it may not be possible to fully reconstruct all the dimensions that wou ld allow for a comparison of standard units. This brings us back to the notion of interaction scale, a ctivity scale, and systemic scale; and, the interactions those the scales are b ased on include person-material, person-person, and potentially materi al-material interactions. However, ANT reminds that interactions are not “isotop ic” in that all the impetuses of action are not necessarily present at the loc us of the action, not “synchronic” in the sense that what is co-present at the locus of action may have begun acting at different times (including a past which is beyond the discursive consciousness of sentient actors), not “synoptic” in the sense that not all of the participants in the interaction are visible to all the other participants, not “homogenous” in the sense that multiple categories of people (e.g. living and dead, present or remote) and things (i.e. multiple t ypes of materials), and not “isobaric” in the sense that different participants may exert different pressures on


60 the interaction (Latour 2005:200-201). ANT provides procedures for examining how the local articulates with the global; for example when assessing how individual interactions articulate with the conceptual core of an activity, applying an “oligopticon” means to limit observations to those interactions that connect an individual node to other nodes in the network (Lato ur 2005:181). However, interactions may well shift through multiple localities before they can be combined into an activity. Moreover, activitie s may shift through multiple locations before they can be viewed together as part of a system (Rapoport 1990). Thus, when we bind time and spatial movement to bring more of the participants into view, what we get is not a demystifie d view of a unified structure, but a “panorama” that coheres a specific moment in the o ngoing system of activities (Latour 2005:187). What is most important is not how much of the total system can be brought into view, but rather the number of connections that is shown by temporarily taking a wider viewpoint. Conve rsely, we must also be aware, at each interaction, of how actors are different iating themselves from the generalizing effects of institutional engagements; thi s is the variation of practice, by agents who have variable received views of their ow n connections with institutions, in what ANT would call “subjectifiers, perso nnalizers, or individualizers (or plug-ins)” (Latour 2005:207). ANT is, in reality, not so much a theory, as it is a se t of methodological propositions that attempt to resist being overly constrai ned by theoretical commitments; its strength is in providing a rationale f or not using theory to predefine aspects of a research universe, but it does no t provide much guidance


61 for understanding the thought processes behind the actor s’ actions. For that, we turn to the concept of the “semiosphere,” as the cognitiv e space within which “it [is] possible for communicative processes and the creation of new information to be realised” (Lotman 2005:207). We also see the netw ork shape in the recognition of sign systems, at both the level of culture and the individual, with a “cultural nucleus” of core institutional attachments conn ected to a “peripheral space” that engenders a “semiotic individuality” that al lows the differential adoption of non-core institutional attachments (Lotma n 2005:209,212-213). “The semiosphere has a diachronic depth, since it is allot ted by virtue of a complex memory system without which it cannot function;” this is what allows for both longer-term stability of core “mental-ideologi cal structures” to persist, and for variable idiosyncratic engagement with peripheral idea s or resistance to the core ideas that provide the basis for changer of the long-t erm (Lotman 2005:215,219). This “internal diversity” allows for the group interpr etation of relationships between signs and what they signify to be contested as a “dialogue” between “substructures” that nested into a semiotic hierarchy, with relationship of “vertical isomorphism” to a higher level of cognitive domain (Lo tman 2005:216,218). The highest levels of cognitive domains are themselves differentiated by “five main modes of equivalence: perceptible [i.e. per ceived by the senses], functional, affective, nominal, and by fiat, that is, arbitrary” (Rapoport 1982:119). However, it is at the level of sign-signified relation ships that changes are made, by the virtue of the interpretant; while the sign ma y be iconic (“based on formal resemblance”), indexical (based on “some kind of existenti al relation with their


62 referent”), or symbolic (an arbitrary connection betwee n sign and signified), it is the fluidity of establishing new interpretants that al lows for variation at the level of replicating sign-signified relationships (Preucel and Bau er 2001:89-90). The crucial aspect of the sign-signified relationship is that the changes within the semiosphere, whether verbal or merely ideational, do not necessarily occur at the same rate. This means that no mental state can be taken as a synchronic representation of the thinker’s reality, rather “[t]he structural heterogeneity of semiotic space creates reserves of dynamic processes and repr esents one of the mechanisms for the creation of new information inside t he sphere” (Lotman 2005:214). So, to contend with multiple levels of signs-symbols, chang ing at differing rates, we must posit more than one way to perceive real ity, as multiple ontological orders. To that end, we might accept hier archy of ontology subsumed under the unifying category of a “ethological order; based in a realist ontology [that] implies a research strategy taking accoun t of the fact that all human activities have a biological/genetic dimension;” b ut within that cognitive domain, there exists an “operational order” that pres ents itself in ways that are accessible to quantitative, functional analyses that do no t depend on verbal or textual declarations, a “discursive order” that does con sider how people’s reality is engendered by “the communicative interaction of an a rray of people using language and language-like types of codes,” and a “ico nic order” that “concerns iconic codes and their use in the visionary building of p ossible worlds…beyond language and alien to verbal communication” (Aijmer 2000:4; 2003:3). This


63 conceptual step precedes the actors’ perception of a “uni tary text,” where “meaning does not reside in artifacts or people but in the moment of interaction between the two; and, “symbols’ meanings do not exist o utside of the moment in which people apprehend them and assemble them into mea ningful formations” (Lotman 2005:218; Robb 1998:337-338). Of course, this line of thinking further problematizes t he researcher’s selection of “analytical scales,” since the scales we select ine vitably “determine the characteristics of the patterns inferred” from our co llected data; moreover, the “lived scales” of past actors occurred at multiple, and mul tiply overlapping, spatio-temporal scales (Burger and Todd 2006:236; Lock a nd Molyneaux 2006:8). Thus, the incorporation of GIS analysis does “facilitate an almost effortless integration and display of multi-scale data and aggreg ation of areal units;” but, we must also remember that when we bind time and space with a GIS, we are simply creating various panoramas of possibility, or focusi ng an oligopticon to follow specific connections. Moreover, while a GIS can b e made to reflect both the vertical and horizontal dimensions of recovered da ta, representation of palimpsest deposits may be difficult to realize, even w hen the data are separated into different layers. This requires that spatial scales be reported in terms of absolute units based on the metrics of recovery, along wi th comparisons of patterns at different scales with a rationale for what makes the comparison meaningful; however, we must also accept that this requi res the “critical procedure of suspension preservation and transcendence ” that allows us to take analysis at one level as far as possible, and hold th at understanding while


64 also allowing us to shift to other scales to follow meani ngful patterns (Marquardt 1992:108). The major difficulty with that approach, assuming that researchers present their procedural steps transparently, is the ex pectation that we will produce coherent, unified narratives that represent th e totality of past reality; but, “there can be no single unified history of the world o r even of some restricted place or period, only a multiscalar history written fro m many different points of view” (Bailey 2008:24). The concept of scale, in both space and time, does not l end itself to convenient, universal classificatory units. Cartographic scales, expressed as a ratio of metric map units to metric units of physical space do offer a definite distance, but also rely on generalizing abstraction. D efining a predetermined extent of an area to be studied, a uniform sampling size as a grain that can be related back to the extent, and an intensity of sample, is another means of expressing the researcher’s analytical scale; however, rega rdless of which scales are used to make inferences about perceived patterns, it seems unlikely that patterns associated with people’s associations with particul ar institutions will be coterminous for every institutional attachment. This h as led to the perception that cultures can be neatly bracketed, in terms of time and space; an individual researcher may relate to such conceptions as simplification s that convey a general sense of boundary, but those classifications become more inveterately rooted with ongoing research, eventually becoming fix ed in the wider public’s understanding of the past. Furthermore, there is the danger of making definitions so general that they cease to be useful, such as definin g a region as “smaller that


65 half a continent but larger than a community and its immediate catchment area” (Kowalewski 1997:291); or, so specific that an overabund ance of categories must be created to deal with differences non-meaningful dif ferences, such as defining a hamlet as “five or more houses, none of which is more than two-tenths of a mile from another” (Aiken 1985:394). From an ANT perspective, every individual is being indi vidualized by any number of attachments, at both the macroand micro-sca le; we can deploy an oligopticon to examine all associations in a geographic a rea, or we can deploy an oligopticon to investigate attachments to particular inst itutions, but neither will lead to a firmly bounded locale. With the locations under consideration here, there are a number of records that allow groups and an ti-groups to define the settlement boundaries; however, as research moves outward there can be no expectation that such records will surface to define othe r settlements. For the urban neighborhoods, it is reasonable to equate house w ith household; however, for the rural areas of the study, a household may not be “coresident in a dwelling or residential compound,” although the members of an e xtended household are likely to reside in a dwelling that is situated on prop erty with a single legallydefined owner (Blanton 1994:5). Households, in these cases, may include not only consanguine and affinal kin, but also borders and h ired laborers that reside within the bounds of the household, whether farm land or urban dwelling (although apartments or other multi-family dwellings where there is no indication of intention for form a collective unit, should be tr eated separately). Unless otherwise stated, the term region is meant to entail 1 2 states: Alabama, Arkansas,


66 Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, and Virginia: any area be yond a settlement, but somehow differentiated from the region as a whole, wo uld be term an infra-region. It is assumed that a settlement is, at the level of bein g coresidents of a collectively negotiated space, a community; the definit ion of community transcends both temporal and spatial boundaries. Specif ically, to provide a comprehensive definition: By community [we] mean small groups, such as households, bands, or tribal organisations, but also imagined groupings that may never meet yet hold some interests in common. Communities may be embe dded one within another, overlap, and differ in importance, d uration, interests, and internal structure. Their borders may be firm or por ous (Gudeman 2005:95). Thus, settlement coresidents may be members in many diff erent communities, possibly including attachments that supersede those that con nect them to others of the settlement. The key to evaluating such association s is whether they meaning can be transformed by the articulator, as medi ators, or simply transport meaning, unchanged, from another node in the network (Latour 2005:39). In regard to temporal scale, the first step is to acknow ledge that the recovery of archaeological materials is limited by the “ time depth,” to which a given assemblage of materials (including documentary ev idence) can inform us, and also the “time resolution,” in terms of dating t hose materials and inferring temporal changes within a deposit (Bailey 2008:14). F ernand Braudel illustrates


67 the diachronic impacts of history on any moment in time. In his view, changes occur at different temporal scales; the “ longue dure ” is the scale of geological and many biological changes, “ conjonctures ” are the intermediate scale of changes in demography, political boundaries, and economi c patterns, and “vnements” are the scale at which individual interactions occur, as w ell as the scale at which many documentary records are published (Bin tliff 2004:176). The intermediate scale can be further subdivided into conj unctures on the “geographical” scale of the “long-term demographic move ments [and] the changing dimensions of states and empires,” and the shor ter term conjunctures of the “economic” scale of “rates of industrialization, t he fluctuations of state finances, and wars” (Braudel 1995:899). Braudel’s categories are meant to reflect patterns of cha nge over thousands of years, hundreds of years, decades, or the ‘ quick time’ of events as they happened; this has led to questioning the utility of those scales to archaeology, where the pattern in the process of indi vidual activities may not be discernable on the scale of “ethnographic time,” but onl y on the long-term scale of “archaeological time” (Smith 1992:26). Unlike anth ropological work among living populations where the “ethnographer and event are isochronic” (Deetz 1988:16), even if the record of the event is put into a finalized written state after the event has occurred; however, the ultimate pursuit o f archaeology is analyzing changes over long periods of time, and no matter how d iscontinuous, it is a largely record of the accumulation individual events t hat provides the archaeological record. The next question is whether or not the archaeological


68 record is a “distortion of the past,” only “because we d o not accurately understand the relationship between statics and dynamics” (Binford 1981:200); again, this is the reason why the “synchronic statement” m ust be a “prelude to a diachronic explanation,” as the individual interaction communicates only an individual’s personal connection with whatever institut ions have associations an activity system, and because “proceeding to diachrony with out acknowledging synchrony is impossible” (Glassie 1975:8). This study began with the creation of a synchronic depi ction of the structures of the built environment or patterns of pro perty ownership, for particular urban and rural settlements in Hillsborough County. More specifically, it is a statement about the metric distances between stru ctures recorded by static documents, especially between residences and structures that are purported to be of particular importance to the coresidents of each se ttlement. The ideal would be to have structures that were still standing, b ut the vast majority of the original structures are long gone. That makes the use o f historic documents (including oral histories and non-verbal means of docu menting history), and GIS analyses, the only means of evaluating the diachronic de velopment of these settlements. As a study of the overall pattern of human settlements, this is certainly well within the tradition of settlement archaeology; but, we are going a step beyond that, in the assumption that the “position, ar rangement, and orientation” of structures “examined at a range of scales constitutes a form of non-verbal communication, and “that spatial patterning of indivi duals, or of the material


69 component of behavior, is an ancestral trait that must predate the evolution of hominids” (Ashmore 2002:1173; Fletcher 2004:116). Most importantly, the approach of the study allows for the actors of the past, as well as the descendant communities of the present, to have a voice in the const ruction of the narrative about the diachronic changes shown by the documents of th e historic record. In that sense, this does not represent the final report of a completed study; this is simply a necessary step, in an ongoing process that uses hi storic records as companions to the archaeological record, to mitigate th e loss of impermanent architecture, in support of the idea that “the federa l government should recognize this land as African American heritage land” (Mitchell 2 000:578). Settlement Archaeology An effort to understand how small, seemingly isolated (both geographically and socio-culturally) communities integrate into large r regional settlement systems needs a methodological and theoretical organizati on that is capable of analyzing the household as the basic unit of a settlemen t, and the settlement as the basic unit of a regional distributions of loci of i ntensified human material behaviors. The objective is to understand how human se ttlements articulate with each other at differing scales, so a broadly-focused “Sett lement Archaeology” will be used to determine the boundaries of multi-scalar an alyses of human occupations of various landscapes throughout time. Howeve r, studying the distribution, nature, and extent of human settlements has found broad application among archaeologists, so it is necessary to state clearly bo th how settlements have been factored into archaeological interpretations generally and how they


70 will be used within this study. The temporal orientat ion of this study is primarily diachronic, in the sense that it is intended to underst and the nature of settlement changes over time, and to explain which socio-cultural as pects motivate different changes in the way humans organize themselves across the l andscape (Plog 1973). One detail shared in most of the literature regardin g settlement archaeology as a “focused methodological concept of archaeo logical research” is a professed gap, covering most of the first third of th e 20 th century, in any demonstrable archaeological concern for house forms, composi tions, or distributions (Chang 1972:1; Parsons 1972; Trigger 19 67). It seems most typical to credit Julian Steward, not so much for his own work d uring the mid-1930s, but for “suggest[ing] to [Gordon Willey] the lack of, and n ecessity for, settlement pattern studies in archaeology,” in 1945 (Willey 1953: xviii). It is Willey’s Prehistoric Settlement Patterns in the Vir Valley, Per that is generally regarded as the seminal work “concerned with locating and mapping archaeological sites on a regional scale with the express purpose of inferrin g sociological processes from changes in site patterning through time” (1953:x vii), although Robert Braidwood appears to have also used settlement patter n surveys as part of his research in Syria, during the mid-1930s (Parsons 1972:1 28). In later writings, Willey would state that “all-inclu sive settlement pattern study was an attempt to prepare a groundwork for an ar chaeological reconstruction that would approximate a total society, n ot just an elite segment of that society” (1989:170). Chang referred back to “Wi lley’s settlement pattern


71 study,” as being the “first time” anyone had “offered a systematic methodological framework for a ‘conjunctive’ approach” that could be u sed for “cultural reconstruction as well as cultural relationships” (1972:2 ). In this study, we will be adopting the simple levels of analysis of Trigger; specif ically, the 3 levels of analysis regarding settlement pattern studies are: “the individual structure, the settlement, and the settlement distributions,” which we note “includes an inquiry into both the synchronic, or structural, and diachronic, o r developmental aspects” of the social relations that we are studying through set tlement patterns (Trigger 1967:151). Chang adds that, on each of the 3 levels o f analysis, the units of analysis represent ‘nodes’ in a network, and that “a stud y of the interrelationship of (‘network’) of settlement units may be said to shift up and down among these three levels” (1972:6). This study accepts Chang’s definition of component for use in settlement pattern studies, namely “an archaeological entity withi n a continuous space and within a meaningful time period, and also consider th e component as the “basic articulatory unit” (1972:9-10). This study has also ad opted Willey’s use of “’micropattern’ and ‘macropattern,’” as related to def ining cohesive ‘community’ and interrelationships between ‘communities;’ however, here the terminology is applied to settlements rather than ‘communities’ (1989 :173). The choice to use pattern is based on a desire to avoid using structure; ho wever, we retain Chang’s understanding that macropattern involves considering “a larger sphere of time and space—to relate a number of different settlements (from two to all that have existed on earth) in many different ways for differen t objectives” (1967:93-94).


72 Thus, we retain the use of “micro-articulation” when d iscussing how networks are perceived and maintained within individual settlemen ts, and “macro-articulation” when discussing how networks are perceived and maintained beyond the individual settlement (Chang 1967:94). The Built Environment as Non-Verbal Communication Amos Rapoport has contended that the environment, w hile not deterministic, does provide contextual cues that function to reduce the range of possible reactions to external stimuli; in short, a wide variety of environmental cues, not only the built environment, operate on huma ns in their environment, and they constrain culturally appropriate choices (1982 :83). Specifically, since all behavior occurs in some context, and that context is based on meaning, it follows that people behave differently i n different contexts by decoding the available cues for their meaning—and these cues may be in the physical environment (Rapoport 1982:69). Within the overall environment, there are multiple elements that convey meaning; In Rapoport’s approach, the three elements are fixed -feature elements, semifixed-feature elements, nonfixed-feature elements. The fixed-feature elements are comprised of things that either do not chan ge, or change so slowly that those changes often go unnoticed within a human l ifetime, such as the natural environment or durable structures; semifixed-f eature elements correspond to features that change fairly rapidly, such as decorative elements, and the nonfixed-feature elements correspond to things that are continually changing, such as a person moving around an environment (Rapoport 1982:87-


73 101). Of particular interest to this study is the concept that sites closer to the community core, “in most traditional societies,” reflect greater social status within the communities of those societies, whereas in “contempora ry societies like the United States,” the opposite is more often the indicato r of higher status within the society (Rapoport 1982:111). In a chapter entitled “Building and Dwelling,” Tim Dant asserts that: The material form of housing reflects the cultural boun daries between different dwelling activities—working, resting eating, sleeping, bathing, defecating. In this sense, the buil ding contains the customs and conventions of a particular culture as w ell as the people who dwell together and their belongings (199 9:65). It seems that worship and formal learning need to be a dded to the aforementioned list of activities, at least in regards to the spatial arrangement of the buildings that represent the ideological core value s that were explicitly stated at the time the community was founded (i.e. agricultu re, education, and religion); however, the buildings themselves appear to have had a n importance that required centrality in relationship to the surrounding agricultural properties. O.V. Burton was careful not to call for environmental deter minism in looking at the development of rural agricultural communities, but he suggested that there is a need to understand the “size, spatial organization, de gree of urbanization, rate of growth, economic function, voting behavior, and even different ideologies” for a sample of all known types of rural, agricultural settleme nts (2002:648). However, that also entails a willingness to understand the organi zational development of


74 settlements that were initially rural, and only later engulfed by the outward spread of a growing urban center; for example, at both Alex andria, Virginia, and at Augusta, Georgia, that the “communities were formed wi th churches at their center, and that these communities were more likely to survive the encroachment of Euroamerican settlement in the previously peripher al locations” (Joseph 2000:111). David Lee’s work in southeastern Florida indicated a sep arate, but parallel, system of central places expressed with Thiessen polygons ( 1992); presumably this network of nodal centers, through which knowledge-b ased capital and other material and non-material forms of wealth were movin g, extended beyond southeast Florida. However, commercially produced good s were widely available to any area where there was a demand; so, unlike differences in consumer goods distributions which may be useful for settle ments of known composition, and material representations of ideology which may be stronger for uniting all types of settlements as American, the spatial arrangement of community structures may offer insights into differences t hat were not drawn from the discursive consciousness of those who initially settl ed the community. Roland Fletcher has done some intriguing work analyzin g the patterns in the use of community space, viewing the built environment as n on-verbal communication; this research applies Fletcher’s models of settlement grow th to an historic site in rural Hillsborough County Florida, especially attempti ng to find common spatial distances being used in the built environment through the multiple phases of community occupation, particularly as the community shifts to the north (Fletcher


75 1995:26,41-42). At the scale of postbellum African Ame rican sites throughout Florida, other territorial pressures may have been bro ught to bear that affected how people and information moved through the physical world. Bealsville may have had particularly little boundary traffic, and r elatively few interconnections with relatively few nodes beyond the community. Thus, even continuous occupations of space do not indicate the dynamics of social interactions within that space, and people may have more connections of co mmunity with other groups, at a distance, than with those that are nearer to them in space; this would reconcile the idea that African Americans moved acr oss the landscape, between rural and urban areas, and yet still “live in areas so intensely segregated that they are almost completely isolated from other g roups in society” (Mitchell 2000:535). In other words, the distances spanned by mi cro-articulations internal to a community cannot automatically be tallied up and transformed into macroarticulations of distant settlements, in a larger netwo rk. Moreover, buildings such as schools and churches may provide both micro-articulations that bind a settlement together as a community, and macro-articula tions that signal, to outsiders, that the settlement as a whole is a member i n group formations and communities beyond the physical extent of the settlement itself. The point of taking the built environment as a form of non-verbal communication is to allow the setting of human interact ions and communications to “guide, constrain, and limit behavior without bei ng determining” (Rapoport 1982:83). The meaning communicated by each setting, as a context of action, may vary depending on reconfigurations at the level o f semi-fixed features, or


76 depending on how the non-fixed features (i.e. human s) present themselves within the setting; these differences may radically cha nge the meaning of fixed features of the setting, and such variations may not b e readily apparent in the archaeological record. For example, a particular settin g may have been used by one group during the day and another group at night ; however, the palimpsest nature of the archaeological record may compress those two separate uses into one interpretive signature. At the level of fixed f eatures, we are more likely to recover cues that suggest appropriate behaviors for all u ses of a particular location, to “routinize behavior, reserving cognitive channel capacity for more important matters;” although, “before cues can be unde rstood they must be noticed, and after one has both noticed and understood the cues, one must be prepared to obey them” (Rapoport 1982:59,236). Thus, “environments both communicate meanings directly and also aid other forms of meaning, interaction, communication, and coaction;” in fact, the non-verbal messages (i.e. meaning) received from the en vironment is a source of non-human agency (Rapoport 1982:50). Furthermore, t hose messages may not convey the meanings intended by those originally constr ucted the fixed features of the built environment; and, that “inertia of the material framework” can become a source of negative feedback at the levels of the opera tional ontological order, and practical consciousness, long before settlement coresiden ts become aware of spatial disorder at the level of the discursive consciousn ess (Fletcher 2004:130). The built environment, then, is an impor tant source of organizing space, time, communication, and meaning (Rapoport 1982: 178-182). The


77 organization of space is likely to be the most visible, even archaeologically; this is the reason that the space between activity areas is so use ful, because it presents a view of the overall use of the space within a site. The organization of time may be less reliably perceptible; for example, a “major th rough-route” may reduce the density of social interactions on the narrower roads of a neighborhood, and “speed the transmission of information and resources,” by forcing “route differentiation” choices on travelers (Fletcher 1995:13 8-140). From a geohistorical perspective, the routes outside a particular boundary (e.g. settlement or region) may have “conditioned the route s inside the region” (or settlement), according to the long-term levels of traf fic, in people and goods, along a particular route (Kinser 1981:79). The organization meaning and communication may be th e most difficult aspects to apperceive archaeologically, in the sense that discernable patterns, as messages, are more culture specific than patterns that att est to the organization of space and time. Moreover, the messages from the mat erial can be polysemous, conveying “high-level (ideo-technic), middl e-level (socio-technic), and low-level (technomic) meanings;” most importantly, all the instrumental and latent meanings of the three levels must be considered, as “they are complementary rather than conflicting or competing” (R apoport 1982:221-223). It may be easier to establish means of restricting, or damp ening, means of communication, especially where time was influential on the overall organization of communication in an area; for example, an unbroken wall reduces the signal capacity of information from beyond the wall, althoug h extreme noise or a


78 tapping code may be able circumvent those dampening ef fects. Furthermore, personal decorations of non-fixed features, or familia rity that mitigates the need for “locating people one does not know in social space,” may not be visible in the archaeological record at all, and environmental cues are likely to be of less importance in situations where mechanical solidarity rei nforces expectations of individual behavior through commonality of expected b ehaviors among all actors (Rapoport 1982:183-184). In any case, since “[d]iffer ent cultures have different institutions for similar activities,’ the boundary of a setting (e.g. household or settlement) cannot be defined a priori but should be based on all identified activities being “‘distributed’ among settings,” as an op en “system of settings” (Rapoport 1990:16). Material as Behavior Roland Fletcher has suggested that there is a connection between the use of material behavior, including the built environmen t, and the mitigation of interaction stress that may prevent increasing urbanizati on due to lacking a previously existing communication assemblage (1995). Fle tcher has pointed out repeatedly that archaeologists’ explanations of past hu man behaviors will not be truly satisfactory until the discipline has developed its own social theory, and that theory will have to consider the long term effects on d ecision making that are created by their material world (1989; 1995; 2004). This ‘material as behavior’ approach should consider metrological consistencies empiri cally, and consider the possibility that mathematical regularities exist in those metrologies that can contribute to understanding how the more durable mate rials created by humans


79 can affect their decisions over longer spans of time than are revealed by the short timescales of activities and speech acts; thus, the mat erial becomes ‘an actor without intent’ (Fletcher 2004:112). The spat ial messages of the built environment are a “class of slow behaviour that has effe cts over a wider range of timescales than active behaviour;” and, while they are learned, the complete message is not normally in view to an individual human actor, hence the message is partially perceptible and also partly an “ele ctrochemical version of the material message” (Fletcher 1995:20,47). This should no t be taken as a muddled position between humans as biological entities i n the long-term, and social actors in the short-term; but, the nest building habits of higher primates, such as gorillas, does suggest “that spatial patterning, whether of individuals or of the material component of behavior, is an ancestral trait that must predate the evolution of the hominids” (Fletcher 2004:116). The generation and maintenance of such spatial messages f rom the environment does not depend upon actors being conscious of the spacing used, or upon exacting measurements in normal interaction; h owever, “the structures of a settlement act as a communication device transmitting a message about a way in which space can be arranged in the horizontal plane ” (Fletcher 1984:197). The attempts to interpret these non-verbal, material messages have primarily taken two forms, and both are selectionist premises; in sh ort, there is no assumption that deteriorating material conditions, or t he negative effects of material inertia, will be corrected through conscious ch anges, although existing material and social behaviors may be used innovatively to contend with a milieu


80 that does not satisfactorily meet existing needs. The f irst use has been to assess “all the types of spacings we can identify in a settlemen t;” and, once all identifiable measurements have been taken, to look for clustering in measurements of each specific type of feature (e.g. roo m lengths, or doorway widths), and to consider how nonconforming measurements might signal material dissonance as spatial disorder that has a deleterious effect on the normal conduct of social action and verbal meaning (Fletcher 198 4:197; 1991:37-38; 2004:133). The second use has been brought to bear o n the topic of urbanization, as an indicator of interaction-communicat ion stress; the underlying premise of this use is that compact settlements (i.e. ab ove 10 people per hectare) are constrained by the area that can be adequ ately covered by the existing communication assemblage, in an environment of increasing stress from a high density of social interactions (Fletcher 1995). T he interactioncommunication stress model assumes that as settlement “densit y ceiling decreases with an increase in community size” that allows f or “more space per person because they increasingly subject to more interacti on effects which increase exponentially” (Fletcher 1995:71). In the first use, the basis for interpretation is “the simple, hypothetical description of a spatial message [that] provides a succinct paradigm for the relationship between variation and the pattern of ce ntral tendencies in a material message system” (Fletcher 1995:40). In the cases under stu dy here, we do not have sufficient preservation of the built environment, from the period of initial settlement, to analyze spatial utilization for specific types of structures; however,


81 we do have two rural settlements that are similar in e very respect, except for the race of the residents. The existence of a series of maps sh owing the locations of structures, in each of the rural settlements, provides us with a opportunity to study the central tendency and variation in measuremen ts between structures in those communities, especially between domestic structures an d those that were the site of providing group services (i.e. schools) or gr oup rituals (i.e. churches). In the study of urban uses of space, we have the oppor tunity to study the effect of population density on neighborhood development; a nd, in one case, we are able to examine a rural settlement that was absorbed i nto an urban periphery, and compare its spatial signature to that of the comm unities that remained rural. The comparison between the rural settlements does prov ide “equivalent cases which can be consistently defined;” but, the singular d istinction of race, between two settlements formed by connected groups of Baptists, a llows us to “study the relationship between dissimilar, linked entities” (Flet cher 1977:67). The use of the interaction-communication stress model, on neighborho ods within a single urban setting, is somewhat different than the origin al use; however, it should be noted that the social restrictions on relocation did creat e firm boundaries for the neighborhoods, and the model is useful as a means of esti mating interaction stress within crowded neighborhoods, and suggesting a mean s of understanding the ongoing expansion of the original neighborhoods, including their present configurations. Another possible difference in understanding, due to th e novelty of uses for these models, is that there may well be a “direct correlation between variation


82 and population size,” which is not expected in the typ ical uses (Fletcher 1977:79). One factor that might impinge on the normal growth o f African American communities, especially after 1877, might be the loss of land due to poor climatic conditions leading to failure to remit mortgage payme nts, with subsequent loss of property; if new properties cannot be acquired outside the existing settlement, then the only option may be to subdivide properties that have remained under the control of settlement residents, and that would lead to a steady decrease in variation. This study does consider overall differences in the number of times an original property has been subdivided, and the mean size of lots created by such divisions; but, a reduction of variation might provide a different insight, based on structures being built to house people separately, on l and that has not been legally transferred to a new owner. Simply put, th e number of property subdivisions in the rural communities could be similar, ma sking the possibility of informal transfer of property between family member s; thus we must consider both the number of subdivisions, and the reduction of v ariation in the spacing of structures within each community. To some extent, the c entrality of socially important structures depends on visual inspection of th e resulting maps, although the visual perception of centrality should be taken in conjunction with changes in central tendencies. Research Design and Meta-Design Given that a unified narrative is not always an expe cted outcome, it may be prudent to discuss this project in terms of its final o utcomes, and research design. The following sections should not be regarded as a finished project, to


83 be shelved prior to moving on to a new project. In t aking an ANT approach, it would seem disingenuous to suggest that research grants, co ntractually obligated scope of work agreements, and the requiremen ts of academic programs do not create boundaries that constrain what will be included in an individual report; but, each report is simply a new sta ging point for planning the next phase of research, as boundary object between resea rch opportunities and community objectives. The final report will seldom pro vide closure, only a “reflection space” that “communicate and facilitate share d understandings across spatial, temporal, conceptual, or technological gaps” be tween “Communities of Practice” and “Communities of Interest” (Fischer 2000:531 ; 2004:156-157). This is where the concept of meta-design comes to the fore, an d the report becomes a “seed” to share with our research partners in descende nt communities; if our fellow researchers are to become equal contributors, then they must be included before the first phase of “evolutionary growth,” and they should be empowered to share in the direction of “reseeding” rather than just correcting the potential misperceptions of the earliest phases of research (Fischer 2007:197, 202). The truly critical phase of this research cannot begin until this first phase has been approved by an academic committee (as a Communi ty of Practitioners), and subsequently presented to the various descendent co mmunities documented by this phase of the research (as Communities of Interest ). However, those communities cannot be the only ones involved. To becom e truly useful, this program of research must seek out, and engage, similar p rograms of research and similar descendent communities (even those that now only exist in the


84 memories of descendents). The larger questions of how hi storically black settlements, both rural and urban, are patterned across ever vaster territories, to what social and economic extents were people incorporate d into neighboring communities, and to what extent their material enviro nments have been preserved, whether as contemporary built environments o r in maps and primary documents, oral traditions, or only in the archaeologica l record, can only be answered by creating partnerships between descendent, aca demic, and governmental agencies. Our meta-design has to create both “an immersive enviro nment [that] allows stakeholders to become deeply engaged in problem solving in the context of information, action, reflection, and collaboration s relevant to the situation,” and also “an emergent environment [that] addresses the nee d for this context to grow and evolve based on ongoing problem-solving activities (Fischer 2000:533). One of the issues with many studies of African American history is the bias of the literature, and that cannot be allowed to perpetuate an analysis based on “failure studies” (Woodson 1989:1028). The point is not to “hi ghlight the most extreme and unusual aspects…as seen from outside and above,” and report to governmental agencies as paternalistic benefactors to be b rought in to fix problems (Wacquant 1997:342). These areas cannot be “an alyzed wholly in terms of lack and deficiencies (individual or collective) rather than by positively identifying the principles that underlie [their] inte rnal order and govern [their] specific mode of functioning;” for the settlements that still exist, there must be far


85 more right than wrong, for them to have continued to function (Wacquant 1997:342). The advantage, and the burden, of working with recen t spatial patterns is that they can be used to explain, and ameliorate any deleterious effects that are associated with the development of those patterns. Ther e are negative issues, but we must also consider how communities sustain themselv es despite such issues. One such issue that has long been a focus of urban economics is the impact of residential segregation and job decentralizat ion on unemployment. Such studies normally emphasize the use of dissimilarity indices that reflect how many people would need to move from their present lo cation to create an even distribution for a given administrative area, and isol ation indices that reflect how the concentration of a specified group into a particula r subsection of an administrative area (Boustan and Margo 2009; Glaeser a nd Vigdor 2001; Massey 2001). Urban economists frequently do a thorough job of tracking historic changes, but often cover so many administrative areas th at illustrative maps are not feasible (Bailey 1973; Weinberg 2000; Wright 198 7). In a study of 195 metropolitan statistical areas, incl uding 34 with populations over 1 million, it was suggested that “[a] one standard deviation reduction of black centralization would reduce the blac k-white employment gap by 17-22 [percent] for the average individual 18-30” (Weinberg 2000:126). If the dissimilarity indices of those same metropolitan areas w as reduced to almost zero, meaning that the demographics of any subsection in an administrative area reflected the overall demographics, then the black-whit e employment gap could


86 be reduced by roughly half for most age ranges and edu cation levels (Weinberg 2000:126). Tampa did see some reduction in overall r esidential segregation during the 1970s, but the trend slowed in the 1980s (M assey 2001:402). The three main trends are “clustering” which occurs when neigh borhoods coalesce, “centralization” which occurs when a particular populatio n lives primarily in the urban center, and “concentration” into relatively smal l geographic sections of an administrative area (Massey 2001:409). The reasons for a black-white employment gap range from prejudices in hiring polici es, to poor availability of public transportation between residential areas. At th is stage, the goal is to determine the level of clustering, centralization, and concentration of black residential areas, which can then be used as the basis for bringing together government administrators and residents to develop mutu ally agreed upon goals. One of the most promising means of building cooperativ e heritage management plans, between Communities of Practice and Communities of Interest, is to establish potential TCPs before they are threatened with a proposed change. TCPs are all places that are importa nt to at least one person that claims an affiliation between the place and some traditional, cultural value, and that affiliation is reflected by at least one of the following: spiritual power, practice, stories, therapeutic quality, and remembrances ( King 2003:100-105). All of the places be discussed here have in common a mini mum of complete structures that might be eligible as historic properties, but they do a have a place in the memories of living descendants. However, it is not necessary to wait until they are further threatened by development to become the subjects of local


87 ethnographies that serve to establish their “traditiona l cultural significance…derived from the role the property plays i n the community’s historically rooted beliefs, customs, and practices” (Parker and King 1998:1). There may well be cultural properties that have signif icance to a specific group’s traditions that are not likely to be shared with outsi ders until some proposed change would affect that locale, but many properties th at could be included are frequent subjects of local stories, which is one way to i dentify possible TCPs (King 2003:102). In the case of Bealsville, “the patterns of land use reflect the cultural traditions valued by its long term residents” (Parker an d King 1998:1). Moreover, there are numerous cases where property is owned by desce ndants of the community’s founders. In the cases of Willow and urban Tampa, there have been previous archaeological studies which suggest they be eligible on the basis of academic value (Panamerican Consultants 2001; Weism an and Collins 2004a, b). By creating a spatial database of these areas, in a GIS format, we can continue documenting the areas as potential TCPs. The ke y is to tie together all information, spatially, so that as future changes are p roposed they can be met with areas of potential effect for which alternative p lans might be developed. This sort of proactive scoping could prevent needless str ess on those who hold places to be traditionally significant, and reduce need less expenditures on redesigning plans and court appearances when partially completed projects are contested by public stakeholders.


88 Chapter 5: Research Methods Studying built environments that often no longer exi st as standing structures is decidedly the province of archaeology. Howe ver, studying the environment as a means of non-verbal communication open s up a means of connecting the environment of the past and the environ ment of the present, with the continuity of an ongoing discourse. Unfortunately this study does not include subsurface exploration of the archaeological properties within the rural communities under study; however, this is seen as the fi rst step toward filling in enough gaps in the information to allow for minimall y invasive future explorations. The bulk of the information in this study, at least to this point, has been the result of reviewing historic documents regarding early land te nure in eastern Hillsborough County, Florida. Throughout the course of research, many different types of documents have been reviewed, including county tax rolls going back to 1883, United States Census Bureau records, and Florida census records; however the key to understanding the earliest phases of property ownership in the communities of Springhead and Bealsville, stems from the General La nd Office (GLO) records for properties that were actually deeded to settlers i n T29S, R22E. Fortunately, the records are accessible online through the Land Boun dary Information System (LABINS); the same website also supplies the original color plat maps used by


89 GLO available, true color Digitally Orthorectified Qu arter Quads (DOQQs) from 2004, and Digital Raster Graphics (DRGs) of United Stat es Geological Survey (USGS) 7.5’ topographic quad maps (Florida Departmen t of Environmental Protection 2007; Florida Department of Environmental Protection, Division of State Lands, Board of Trustees Land Document System 200 7). Once distances were calculated, consideration was given to normality of the distribution figures; however, since the presented samples are not randomly sel ected, only the Mann-Whitney non-parametric test was considered to compa re the distance found in the rural communities. The normality of dist ance distributions were tested with SPSS; however, all further statistical anal yses were conducted, and associated figures were created, with PAST (Hammer et al 2001). The distances to a theoretical community center, in the preliminary research phase, were calculated based on known residents o f the eastern half of T29S, R22E and were accurate for a basic breakdown of wh ite versus black residents in the area; however, some of the white prop erty owners, while sometimes related to residents determined to be part of the Springhead community, were later determined to have been residen ts of Trapnell, Hopewell, Turkey Creek, or even the southern edge of Plant City (Robinson 1928:81-82). Furthermore, a few of the known founders of the Bealsv ille community were discovered to have been granted deeds to properties in T30S, R22E. The GLO records for Hillsborough County include information abo ut the recipient of the deed for each parcel of land, and also a text descriptio n of where that land was located; using that information, it was possible to geor eference the GLO color


90 plat map to an actual physical location on the Earth’s su rface, and associate those property records with the names of the deed recipi ents. A number of other historical maps were also georeferenced, including county road maps from 1934 and 1958 (Publication of Archival Library and Museum M aterials 2007), and each of those contributed to an understanding of property ownership, and/or changing land uses. Ultimately, due to lack of evidence for whe re actual structures were located, the GLO records primarily served to confirm in formation from other historic sources about the identity of the early settlers of both rural settlements. For 1916, the Dixie Survey Company map was available to show the positions of houses in the rural study area, along wit h names of property owners and the acreage they owned; for later structures, the A rcGIS software was used to digitize the structures shown on the USGS 7.5’ topog raphic quad map (University of Alabama 2007). For the 1947 and 198 4 structures, there were no maps available that showed the locations of residences specifically; therefore, it was decided to digitize the structures shown on the 195 5 USGS 7.5’ Nichols Quad topographic map, with structures from 1947 aerial photographs shown in black and structures from 1984 aerial photographs shown i n purple. The earlier location of the Antioch Baptist Church was confirmed by t he oral testimony of G.S. Cunningham, who was young church member when the building was moved, and the stated location was consistent with the la nd use depicted in the 1938 aerial photographs showing cleared land across from the Pine Hill Cemetery and the apparent trail that was used to move the building over Howell’s Creek to the church’s present location on the west side of Horton Road


91 (Hillsborough County and City of Tampa Water Atlas 20 07). The locations of the Baptist church in Springhead were reconstructed from h istorical maps, historical aerial photographs, contemporary color DOQQs, and conte mporary Hillsborough County property appraiser land ownership records. Due to the restrictions of the license for the available version of ArcGIS, the data w ere exported to Quantum GIS to calculate distance matrices between all structures, a nd between general structures and the proposed socially important structures of churches and schools (Quantum GIS Development Team 2010). The resu lts are reported as both summary statistics, which are used as a convenient in dicator of overall variation in the built environment through time. The locations of the historically black neighborhoods of u rban Tampa were digitized according to the descriptions of the Raper re port; however, the boundaries of Dobyville were defined according to a St. Petersburg Times newspaper article, and the boundaries of Ybor City and the Scrub neighborhoods were defined according to a map in More Than Black: Afro-Cubans in Tampa (Greenbaum 2002:86; Tubbs 2008). All structures in Ta mpa locations were digitized on the basis of Sanborn maps; the maps were a cquired in PDF format, from the Tampa-Hillsborough Public Library, converted to JPG images, and georeferenced within ArcGIS (Tampa-Hillsborough Count y Public Library System 2007). Since some road routes have changed throughout the period under study, and the roads themselves may have been widened to acc ommodate more traffic, the measurement function of ArcGIS was used to ensure t hat the proper scale of the Sanborn maps was maintained. Most importantly, th e Robles Pond

PAGE 100

92 neighborhood is used to represent a community that was originally rural in character, and outside the Tampa city limits; moreover, the community had a documented schoolhouse, and both Baptist and A.M.E. church es, to provide community focal points. In the case of Robles Pond, t he 1922 and 1931 Sanborn maps were used to digitize structures, and as the basis f or distance calculations between domestic structures and the proposed socially impor tant structures. The populations recorded by Dr. Mays, in 1927, were u sed along with the total areas of the respective neighborhood, as a meas ure of historic population densities. For comparison, and to document neighborho od coalescence, the 2000 U.S. census block groups were used, along with dem ographic data, to show the current distribution of African Americans. The demographic data was appended to the spatial data using the official block g roup codes as a primary key. The same information was also utilized at the coun ty level, for the 795 census block groups from the 2000 U.S. Census. For Willow, the structures were digitized based on a ha nd-drawn map from an unpublished manuscript of oral histories that has bee n preserved by the descendants of the mill owners, and from a visual inspecti on of a georeferenced 1938 aerial photograph. The boarding house for sing le men was not used in conducting spatial analyses because it remains uncertain as to whether it was inhabited by both black and white employees. The own ers’ houses were also excluded because some family members appear to have onl y been in Willow on weekends, returning to Tampa during the week so that th e children could attend school there. The commissary, and the various buildings a ssociated with the

PAGE 101

93 operation of the mill would have been socially impor tant buildings, but their location was likely to have been highly affected by access to transportation networks.

PAGE 102

94 Chapter 6: Results The initial settlement of Bealsville remained close t o the tributaries of the North Prong of the Alafia (Figure 5). Of the rural settlements considered here, the level of settlement nucleation was greatest at Bea lsville, and the residents of the town were clearly willing to expend a great deal of time (~3 weeks) to move the existing Antioch Baptist Church, on log rollers (ove r a creek), from where it had ended up on the periphery of the community, to a close approximation of the mid-1920s center of town. Roughly a decade later, mo re community activism would lead to the community’s new schoolhouse being buil t at that same central location (Klugh 2004). There definitely appears to h ave been a consideration of centrality in the placement of the Baptist churches in both communities; however, in Bealsville, the Baptist church occupied a central posit ion early in the community’s history, then found itself on the periphery and was physically moved back to a more central location (Figure 6). Ear ly in the history of both communities, drastic weather conditions placed an extra ordinary strain on the abilities of the residents of both communities to retain possession of their land; the severe winter frosts of the mid-1890s forced many of those residents to sell off their properties and move on to greener pastures, so to speak. The earliest properties of Springhead were also locate d near the tributaries of the North Prong of the Alafia (Figure 7). The Springhead Baptist

PAGE 103

95 Figure 5. GLO Locations of Bealsville Founder Prope rties

PAGE 104

96 Figure 6. Growth of Bealsville’s Built Environment

PAGE 105

97 church was originally on the property of the Howell fa mily, who constitute one of the elite families of that community; however, the po sition of that church was on the western edge of that community (Figure 8). That might reflect a connection with the families of Trapnell. For example, the Spa rkman family had large land holding in both Trapnell and Springhead. In 1948, the property of the Springhead Baptist church was sold to a Methodist church, and a new Baptist church was built on the northeastern edge of that community; how ever, in 1979, the Baptist church was rebuilt toward the center of the community, and that ushered in a period of community growth. In truth, the present c ommunity that is in the location where the Springhead community (which sometim es appears on maps as Springhead Park) was initially sited does not have mu ch of a connection to the original founders of the community, whereas the Bealsv ille location has been a fairly stable community, and many of the current prop erty owners are consanguine kin of the settlers that were originally dee ded the property. The earliest locations of Bealsville’s Antioch Baptist church re main something of a mystery, including a “brush arbor” church that is commonly reported as serving as the earliest church for the community. Interestingly, beyond the impacts of weather, and the ever present land speculators looking to acquire land for either the Seabo ard Air Line Railroad, or various phosphate mining interests, the building of Sta te Road 60 separated the southern third of the Bealsville community from the pr operties to the north. State Road 60 was laid down in the 1920s-1930s, and it appea rs to have hastened the loss of property to the south of that major road. As that property was lost, new

PAGE 106

98 Figure 7. GLO Locations of Springhead Founder Prope rties

PAGE 107

99 Figure 8. Growth of Springhead’s Built Environment

PAGE 108

100 property was gained to the northwest, bringing Bealsvi lle into closer contact with the boundary for Trapnell. For a period of time, t hat left the Antioch Baptist Church on the southern edge of the Bealsville community (Table 1). For Table 1, Table 1. Distance to Baptist Church in Bealsville 1916 1947 1984 Number 22 128 333 Minimum 169.52 54.68 54.68 Maximum 2866.54 2437.16 2460.84 Median 963.48 963.51 1015.35 Mean 1178.73 977.79 1030.56 SD 792.58 460.58 479.20 “number” represents the number of structures measured fr om, and the summary statistics represent distance in meters. In Springhead, th e movement of the Baptist church to the settlement perimeter seems purpose ful, possibly to signal community values to drivers on the road to Mulberry (T able 2). For Table Table 2. Distance to Baptist Church in Springhead 1916 1947 1948 1984 Number 18 213 213 494 Minimum 175.52 133.96 82.13 38.59 Maximum 2702.78 3094.17 4017.43 3621.44 Median 1371.66 1559.49 1949.00 1263.51 Mean 1460.18 1573.95 1940.35 1516.10 SD 844.21 776.40 929.24 911.54 2, “number” represents the number of residences in the community, and the summary statistics represent distance in meters. For Bealsv ille, the earlier location of the Antioch Baptist church would not have f ronted on the newly created State Road. However, if that had been desire d, then it would have been

PAGE 109

101 easier to move in that direction than over Howell’s Cr eek and farther north. This seems to support that the buildings were socially impor tant for both communities; but, in Springhead it was used to communicate to people outside the community, and in Bealsville it was used to communicate community co hesion to settlement residents. Since residential locations were only confirmed for the 1916 community extents, those are the only structures where the names of property owners are known. However, the spatial relationships of all struct ures to the Baptist churches, for each community, were analyzed through the Quantum GIS software, and the PAST statistical package. The mean dist ances of each community’s residence, to their respective Baptist churches, were compared statistically. Due to the small, non-randomly selected sa mple size, they were, once again, compared using a Mann-Whitney U test. With the adjustments made to exclude settlers that were later attributed to oth er communities, to include more Bealsville founders, and with residences being comp ared to known structures for each community’s focal point, the results w ere much different. Moreover, with residents definitely assigned to specific c ommunities only distances to the individual community’s church was measur e, instead of considering distances to each separately. The results indicated that the 2 communities were not st atistically different, at least in terms of the 1916 mean distances between r esidence and local Baptist church (for 1916, U=158 p=0.286), but they are diff erent for the 1947 and 1984 distances (for 1947, U=7570, p=<0.001; for 1984, U=59 293, p=<0.001). The two

PAGE 110

102 settlements were also distinct in terms of distance to scho ol, in both 1947 (U=11351, p=0.010) and 1984 (U=64361, p=<0.001), wh ich are the two dates for which the school locations are definite (Table 3). For Table 3, the “number” indicates the number of structures from which measureme nts were taken, and the summary statistics represent distance in meters. Howeve r, there do seem to Table 3. Bealsville and Springhead Distances to Sch ools Bealsville Distances Springhead Distances 1947 1984 1947 1984 Number 128 333 213 494 Minimum 102.02 102.02 35.48 35.48 Maximum 2461.27 2482.64 2807.32 2838.59 Median 1005.02 1018.99 1155.05 1263.28 Mean 1023.14 1062.38 1197.00 1323.29 SD 503.16 520.91 577.63 672.21 be some consistencies in residence locations at both commu nities; specifically, the majority of the 1916 residences were placed at or above the 100-foot contour interval indicated on the 1955 USGS 7.5’ topographic quad map, and they also seem to indicate a preference for “somewhat poorly drai ned soils” (Figure 9). For the Bealsville community, there does seem to be a tend ency to place houses closer to the dominant linear water features of Howell ’s Branch of the North Prong of the Alafia River, and Howell’s Creek; and, a lthough the residence sitings are still at or above the 100-foot contour, th ere is a greater likelihood that the house was situated over “poorly drained” soils. The population densities of both rural communities are below the 10 people per hectare density threshold, where communicati on systems might limit the growth of the settlement, although the distances b etween all structures show

PAGE 111

103 Figure 9. Soil Drainages for Bealsville and Springh ead

PAGE 112

104 change over time (Table 4, Table 5). For Tables 4 an d 5, “number” reflects the number of measurements from structure to structure, and the summary statistics represent distance in meters. Figure 10 depicts the comp arative development trajectory for both communities, according to measureme nts of distance between every structure. For 1916, the mean distances between structures are almost Table 4. Bealsville Structure to Structure Distance s Bealsville 1916 1947 1984 Number 462 16256 110556 Minimum 92.06 19.23 19.23 Maximum 3824.35 3824.92 3986.07 Median 1318.56 1238.01 1308.80 Mean 1493.17 1297.45 1368.80 SD 796.00 662.10 705.87 Table 5. Springhead Structure to Structure Distance s Springhead 1916 1947 1984 Number 306 45156 243542 Minimum 114.88 19.15 18.25 Maximum 3068.27 4440.10 4655.76 Median 1481.29 1606.24 1757.45 Mean 1515.39 1664.51 1821.80 SD 677.15 835.10 938.58 the same, and there is actually less variation between structures in the Springhead settlement; however, in 1947 the mean dist ances in Bealsvile are roughly 78 percent of those in Springhead, and in 198 4 the mean distances in Bealsville are about 75 percent of those in Springhead Similarly, the variation in distance between structures, in terms of standard devi ation, in 1947 Bealsville is about 79 percent of that in Springhead, and in 1984 it is about 75 percent.

PAGE 113

105 Figure 10. Comparison of Structure to Structure Dis tances Interestingly, both settlements show a similar increase i n distance between structures; however, in Bealsville, it appears that struc tures were being built in between existing structures, creating a new starting poin t before the community started to spread out again. This may well reflect a point in time where the community was not able to acquire new property at its o uter boundary, although normal outward growth does appear to have resumed. The data regarding the number of property subdivision s were also gleaned from the Hillsborough County property appraiser’s recor ds for the current, through 2008, properties within the boundaries of the properties that were originally deeded to the founders of the Bealsville a nd Springhead communities (Figure 11). The only comparison made herein is betwe en those original

PAGE 114

106 property boundaries and the current number of separat e tracts that cross a given property. As a consequence, it is possible that properti es being considered were aggregated and disaggregated numerous times, during t he intervening years. For example, it can be seen on the 1916 Dixie Survey map that the lands originally deeded to Bryant Horton and Elias Dexter were consolidated into the holdings of the Coronet Phosphate Company; later, ma ny of those same lands were reacquired by the wealthier Bealsville residents, such as Alfred Beal and O.V. Hargrett. It is precisely that pattern of reacqui ring lost properties, and keeping them within the Bealsville community that form ed the basis of renaming the settlement, in 1923, to honor Alfred Beal’s part in securing a continued existence as unified community. However, it should be r emembered that what is important here is how the cycle of property loss and rea cquisition affected the size of available properties across the settlement, and the position of new residences. Subdivisions were recorded by number, without differen tiating the size of the component properties. However, along the northe rn boundary of the Springhead community and the southern boundary of the Bealsville community, several properties were actually integrated into large r parcels rather than subdivided. Generally, in the cases of both communiti es, properties that were integrated into larger properties were never reacquir ed from phosphate mining interests, and they are currently either still held by phosphate mining companies

PAGE 115

107 Figure 11. Subdivision of Founder Properties Throug h 2008

PAGE 116

108 or are being used as agricultural land outside the prese nt-day community of Bealsville. The properties that were integrated into larger parcels were considered to have been removed from their respective co mmunities, for purposes of comparing the extent to which the original properties were divided. For the Springhead community, given a hypothetical eq ual division of all properties, there would be an expected lot of around 2.48 acres. However, the former property of Joseph Howell was sold to a planne d subdivision; and, if that property is removed from consideration, then an expect ation of equal division would yield an expectation of 3.11 acre lots. For th e Bealsville community, given a hypothetical equal division of properties, there wou ld be an expected lot of 3.9 acres. However, if the properties that were integrat ed into larger parcels are removed from consideration, then there would be an e xpected lot of 3.32 acres. The calculation of divided acreage is based on the actual divisions of the original GLO deeded properties. When corrected for properties that are no longer within their respective communities, they seem to indicate that the original Bealsville properties were less divided than the Springhead prop erties. Tables 6 and 7 show that name of the person to whom properties were i nitially deeded, the number of acres they were deeded, the number of sepa rate properties on that land in 2008, and the mean acreage based on the origi nal acres and total number of subdivisions. One consequence of dividing farmland into smaller resi dential lots, especially given the poor soil drainage conditions in t he area, has been an increased prevalence in flooding following any substanti al rain; since Bealsville’s

PAGE 117

109 general elevation is lower than Springhead, it has af fected the former community to a greater extent. The reported flooding is attri butable to the poor drainage of Table 6. Property Dispositions of Bealsville Founde rs Original Deed Original Acres Subdivisions Mean Acreage Allen, Eva 120 71 1.69 Beal, Alfred 160 79 2.03 Berry, Isaac 120 0 Consolidated Clyatt, Bacchus 80 2 40 Dexter, Elias 80 31 2.58 Dexter, John 80 47 1.70 Dexter, Riley M. 80 20 4 Glover, William 40 0 Consolidated Hayes, Edmund 80 2 40 Henry, Neptune 80 33 2.42 Holloman, Mills 160 42 3.81 Horton, Bryant 80 51 1.57 Horton, Samuel 40 6 6.67 Segenger, Abraham 40 0 Consolidated Simmons, Newton 120 3 40 Smith, Roger 80 12 6.67 Stephens, Bryant 80 6 13.33 Stephens, Rosa 40 5 8 Story, Robert 40 0 Consolidated the soils; however, before the properties were divide d into so many house lots and before there were so many paved roads running thr ough the area, local residents Andrew Williams and Samuel Berry had been re sponsible for deciding when it was time to manually dig drainage ditches thr ough the farmland, to drain into the natural linear water features of the area (Brown and Brown 2003:269270). Unfortunately, it has not been possible to docu ment the flooding, so that Hillsborough County could be informed of the worst ar eas for flooding and develop an action plan for mitigating the local floo ding problems. It should not come as any great surprise to Hillsborough County that t here is flooding in the

PAGE 118

110 Table 7. Property Dispositions of Springhead Founde rs Origi nal Deed Original Acres Subdivisions Mean Acreage Blackwell, Charles H. 40 6 6.67 Bryan, Nathaniel C. 280 43 6.51 Clemons, William 80 27 2.96 Devane, Mary 160 62 2.58 English, Andrew 120 21 5.71 English, Cornelius 40 2 20.00 English, William 160 69 2.32 Harrell, Arrindy 40 3 13.33 Harrell, Benjamin T. 40 4 10.00 Howell, Cornelius R. 160 35 4.57 Howell, Joseph 200 271 0.74 Howell, Sarah 160 40 4.00 Howell, T. J. 200 91 2.20 McClelland, Silas 40 5 8.00 Morgan, Daniel R. 160 88 1.82 Ring, William H. 160 53 3.02 Sears, Hiram 160 99 1.62 Sistrunk, Moses H. 160 87 1.84 Sistrunk, William H. 80 9 8.89 Sparkman, Bird 80 3 26.67 area, not only due to recent changes in land use, as the y first noted that the area was prone to drainage issues in a 1941 County Land Use P lanning report (Hillsborough County 1941:56-57). The urban Tampa neighborhood of Robles Pond presents an interesting contrast to the Bealsville settlement, although the posi tion of proposed socially important structures is not definite for the period w hen it was a rural settlement well outside the boundary of Tampa, in the late 19 th century. Table 8 shows the “number” of residences in the Robles Pond neighborhoo d for 1922 and 1931, and the summary statistics represent distance in meters. R obles Pond was still outside the boundary of Tampa proper, although the t own had

PAGE 119

111 Table 8. Robles Pond Distances to Community Structu res Robles Pond School AME Church Baptist Church 1922 1931 1922 1931 1922 1931 Number 55 139 55 139 55 139 Minimum 27.80 19.79 47.01 12.81 43.83 12.22 Maximum 313.49 233.69 418.83 339.89 410.25 330.62 Median 147.57 126.16 241.40 177.79 305.85 157.44 Mean 160.96 128.37 243.26 168.48 282.53 157.52 SD 76.79 48.72 94.88 77.85 91.41 74.74 grown out well beyond the settlement; the 1922 Sanb orn map is actually a supplemental volume that documents the growth of Tampa ’s built environment, and the need to determine fire hazards in the growin g town. In 1922, there were 55 residences in the Robles Park neighborhood; and, by 1931, the number of domiciles had grown to 139 (Figure 12). The most stri king change is in the positioning of what is, here, being termed socially im portant structures; in this case, a school, a Baptist church, and an A.M.E. church. Th e school was toward the center of the community in 1922, but the churche s were both at the boundary of the neighborhood; as in Springhead, this may mean that signaling community values to those outside the settlement was more importa nt than centrality to the early residents of Robles Pond. Still, the school was m oved to an even more central location during the intervening decade, and b oth churches were either moved or rebuilt in more central positions; in all 3 ca ses, the mean distance and standard deviation of residence to church or school was r educed, due to both residential in-filling and structure repositioning. Using residential density as a measure of interaction-comm unication stress was a bit more problematic. Although opportuniti es to live outside the

PAGE 120

112 Figure 12. Robles Pond in 1922 and 1931

PAGE 121

113 eight neighborhoods were restricted, at least 200 black T ampans lived outside those neighborhoods. Moreover, while none of the nei ghborhoods exceeded postulated interaction limits as a whole, it is not clear how a bounded neighborhood might deal with such stresses (as opposed to discrete settlements). As an individual block within a neighborhood, the 44 Q uarters reached 518 people per hectare (0.41 hectares in area), which wou ld put it near the interaction limit. There is no indication of how that borderlin e stress manifested explicitly negative living conditions, though it was eventually de molished and replaced by a Catholic church. Looking at the contemporary population demographics of the 2000 census, it is clear that residential segregation is still an issue in Tampa; the 2000 census lists Tampa as having a population that is 26.1 percent African American, and only 112 out of 336 census block groups have at least tha t composition (Figure 13). The neighborhoods that exceed the total populat ion, by 2-3 times the overall percentage, appear to have developed by aggregation of neighborhoods that bordered the original eight neighborhoods listed by Dr. Mays. The data from Ybor City should be excluded because the population de nsity is misleading, as the demographics of that neighborhood would have inclu ded many Cubans that were not counted in the population estimates (Table 9 ). There is a similar level of residential concentration for Hillsborough County, although the countywide African American population was 14.8 percent of the ove rall population in 2000. Only 205, out of the total of 795 in Hillsborough Co unty, block groups were occupied by at least the expected average number of Afr ican American

PAGE 122

114 Figure 13. Black Residential Centralization in Tampa (2000)

PAGE 123

115 (Figure 14). A quick estimate of the general level of isolation can be ascertained by deducting the overall percentage from the actual pe rcentage of African American in any give concentration of block groups. Table 9. Neighborhood Population Densities in Tampa (1927) Neighborhood Pop. (1927) Hectares Density College Hill 4094 118.81 34 Dobyville / West Hyde Park 2835 66.21 43 Garrison 812 26.49 31 Robles Pond 315 11.65 27 The Scrubb 8362 33.36 251 West Palm Avenue 2478 12.63 196 West Tampa 3331 67.92 49 Ybor City 896 148.83 6 The results from Willow are quite dissimilar, due to only the white workers traveling away from the town to attend churches and/o r school. In both cases, the distances are reported in meters, but it may be mo re convenient to convert the figures to kilometers for the white workers. Table 10 shows the distances of black residences, in meters, to the church and school provide d by the lumber company, while the distances for white residences are simp ly the distance to Wimauma. This does suggest that, overall, macro-articula tions with other local Table 10. Distances to Group Services for Willow Willow School Church Wimauma Number 40 40 32 Minimum 62.94 47.77 7888.32 Maximum 529.21 524.00 8355.90 Median 189.95 180.90 8011.76 Mean 246.41 242.36 8037.45 SD 151.22 151.32 126.71

PAGE 124

116 Figure 14. Black Residential Concentration in Hills borough County (2000)

PAGE 125

117 communities were more extensive among the white employ ees of Willow. However, since there do not appear to have been any o fficial restrictions on black employees movements away from company property, it is entirely possible that there were macro-articulations with other African American communities.

PAGE 126

118 Chapter 7: Discussion The level of nucleation at Bealsville does seem incongr uent with the dominant pattern of the Cotton Belt throughout the agricultural South. Bealsville was even more nucleated than comparable Florida sites li ke the Zion settlement in Marion County, Florida. The white settlers of Spr inghead also worshipped together, and sent their children to school together; h owever, their settlement pattern was clearly more dispersed than seen at Bealsvil le. There are some consistencies between Bealsville and reported settlement nucleation in other sugarcane producing areas; this issue must be addressed, so that an adequate regional distinction can be made between Florida and o ther cane-producing regions of the South. The obvious community-oriented structures, such as the Antioch Baptist Church or the Historic Glover School, app ear to have represented ideational attitudes toward the place of C hristian religion and public education in the development of their community; the centrality of the buildings, relative to the surrounding agricultural demesnes, see ms to have been important enough for a small, agricultural community to use a few weeks worth of the community’s aggregate “leisure” time (although they di d play several baseball games over those few weeks, so they didn’t entirely aban don leisure) (Klugh 2004:125). Based on comparisons with the Robles Park nei ghborhood, the placement of socially important structures, during and im mediately following the

PAGE 127

119 Reconstruction, may have been based more on signaling p articipation in American life generally; however, following the incept ion of Jim Crow laws, and their subsequent entrenchment and expansion, African Am erican communities that were able to expend the resources may have seen t hose structures as more important to providing intra-community focal points. One of the saving graces for the Bealsville community se ems to have been its isolation, more than its geographic territory circumscription from surrounding water features, and unusable scrub lands. T he isolation factor, in conjunction with a reliance on subsistence-level truck veg etable agriculture, allowed the original settlers of Bealsville to survive t hrough lean years. They also required the contribution of a “homegrown” bourgeoi s class to ameliorate living conditions for the poorer members of the community, and even to preserve the contemporary existence of the community; in the earliest phase of settlement, the Bealsville settlers did benefit from initial neo-pater nalistic ties, from their slaveryera owners-turned-benevolent caretakers, at least until selected properties had been surveyed and begun to show a return on labor in vestment. Ultimately, the conditions that prevailed at Bealsville did not force t he geographically circumscribed residents to abandon the site of their commun ity; although, negative environmental impacts on crop production, the discovery of pebble phosphate, and in the case of Bacchus Clyatt the decision to move out of the area, did contribute to the changing configuration o f Bealsville’s built environment. Even more remarkably, much of the land is still in the hands of the descendants of the original settlers, many of whom still live on t he same land that was

PAGE 128

120 originally settled by their direct ancestors (e.g. the recently deceased Dr. Sam Horton, former president of the local chapter of the NAACP, owned the land that was deeded to his grandfather, and his great-grandfath er). Instead of the more commonly asserted “wanderlust,” that allegedly preceded the mass-scale migrations of African Americans from the South, at Bea lsville we see a commitment to band together as a unified community, o ne that was fairly selfcontained until well into the 20 th century. Still, the contacts between the two rural communities appear to have been based on negotiating sh ared use of boundary space (e.g. springs as swimming holes); in terms of populat ion movements, the closest options for Bealsville residents were toward conta cts in the highly segregated neighborhoods of urban Tampa, or toward ot her rural African American sites. In regard to the division of properties, from foundin g to the present, eastern Hillsborough County does not provide evidence t hat African Americans subdivided property due to an inability to acquire ne w property through purchase. However, the existence of Springhead to the immediat e northeast, and Trapnell to the immediate northwest, would have made southward the only direction open to regular outward expansion; while there were no nu cleated settlements to the immediate south of Bealsville, the 1916 Dixie Survey map shows that phosphate companies had a strong interest in that area, and thos e properties may not have been available in the early decades of the 20 th century (Figure 15). The change in variation of measurements, between 1916 and 1947, suggests that there was

PAGE 129

121 a significant change in the number of structures on each property; in Bealsville, there is a drop in both central tendency and variation from 1916 to 1947, before Figure 15. 1916 Dixie Survey Map Showing Property O wners returning to what appears to be a normal growth traj ectory. As this research progresses, it will be valuable to get insight from comm unity members regarding

PAGE 130

122 the informal allocation of property. If there was no formal change in documented ownership, then the properties would appear to have r emained aggregated in primary documents, such as tax rolls; so, there may have also been an informal division of other assets, as resources were pooled betwee n property coresidents, to make payments to external governmental entities. T his would be difficult to ascertain through archaeological evidence alone, and wi ll depend on personal recollections. There are a few broad trends in the transfer of pro perties that are made clear through the available historic maps. Through the 1880s, only the Plant Investment Company, the Central & Peninsular Railroad and the Florida Internal Improvement Fund held title to the land around the founders of Bealsville and Springhead. By 1916, the Euro-American settlers in t he southeastern quadrant of T29S, R22E had sold off their properties to a vari ety of new owners, including some private purchasers who became part of the Bealsvill e settlement. However, there was also a complete turnover of non-resident inv estors, with the Virginia Trust Company, the Barlow Land Company, the Coronet Phosphate Company, and the Polk County Phosphate Company acquiring vast tr acts of land in and around Bealsville and a few small tracts in and around Springhead. The owner of a turpentine still in nearby Lithia, Norman Smith acquired 380 acres of land in and around Bealsville (Hillsborough County 1998:74). Interestingly, the 1916 Dixie Survey map shows that Alfred Beal’s property was d ivided among other members of the Beal family (the Holloman and Horton parcels were transferred to married Beal women), and the Virginia Trust Compa ny. On the other hand,

PAGE 131

123 property that had been held by the Florida Internal Improvement Fund, the Sistrunk family (Springhead founders), and Mills Hollom an’s land was acquired in 20 and 40 parcels by Bealsville founding families like H orton, Glover, and Dexter. The Coronet Phosphate Company seems to have acquired most of its land from the Florida Internal Improvement Fund and Plant Investment Company, although it also purchased the northern half (40 acres) of Bryant Stephens and a small portion of the property that was originally dee ded to Eva Allen (13 acres). The Virginia Trust Company purchased the majority of the property that had originally been granted to white owners in the southe astern quadrant of T29S, R22E, acquiring only 35 acres from Beal and 40 acres of the property that originally deeded to Newton Simmons. Of the 120 acre s that Simmons had been deeded, only 39 acres remained, but 20 acres had been acquired by Bealsville residents O.V. Hargrett and William Glover. Norman S mith almost exclusively purchased property that was originally held by the Ce ntral & Peninsular Railroad, except for the 40 acre parcel that had originally been deeded to Charles Blackwell. In urban Tampa, the overlap between the neighborhoo ds in 1927, and the present clustering of African Americans, clearly shows that coalescence has occurred. Furthermore, Tampa shows a significant level of black residential centralization and concentration. Certainly where t here are fewer African Americans than the city-wide percentage, there is not th e same level of segregation that was prevalent in 1927. However, for the residents in those neighborhoods, there is likely to be a great deal of t ime lost to commuting to and

PAGE 132

124 from work, as well as to areas of commerce. The level of isolation, in terms of interactions with people who reported themselves as othe r than African American in the federal census of 2000, appears to be the more d efinite issue. Future research will have to address that concentration, in te rms of access to housing, and transportation to commercial centers and available jobs. The population densities of those neighborhoods, as a w hole, do not appear to have been unsustainable. Still, there was a lso a substantial amount of variation between blocks, with places like the 44 Quarte rs part of “the Scrub” coming close to the range of interaction densities that would be noticeably problematic. It is also interesting to note that only Dobyville has received any notice as a neighborhood with historic value to Tampa; there is currently a marker at the corner of Willow Avenue and Platt Stree t acknowledging the historic value of the neighborhood, and some descendants of earlier residents still live there. The overall demographics of the nei ghborhood have changed, and it now blends seamlessly with the rest of Hyde Park. The value of using a GIS to gather data from various sources, and organize them spatially, is readily apparent. Furthe rmore, the comparison between Robles Pond and the rural sites of eastern Hill sborough County, suggests that communities will muster resources to move, o r rebuild, structures that do not meet the needs of the community as whole. Although, more sites must be included before we will be able to determine whether that is a matter of spatial needs, or socio-cultural needs. From recent censu s records, it does not appear that there is a significant concentration of Af rican Americans in the vicinity

PAGE 133

125 of Willow. In Bealsville, there is an extant communi ty with historical ties to the area that is already engaged in historic preservation. In urban Tampa, there is a significant concentration of African Americans, with nei ghborhood clusters having expanded from historically black neighborhoods. In terms of what would qualify these properties as TCP s, it seems clear that stories and remembrances are what primarily delin eate the boundaries of these settlements. For the two rural communities in ea stern Hillsborough County, there is not much of a perceptible difference to the o utsider passing through. The things that distinguish the communities are the affe ctive, nominal, and arbitrary cognitive domains. A local resident likely pe rceives indexical interpretants that communicate boundary information f rom geographic features like streams and perhaps man-made features such as roads th at are part of the operational and discursive ontological orders. However, it is the iconic order that allows for a church to communicate the values of a commu nity to those who are not part of the community, although the interpretan t is symbolic rather than iconic. For the Willow site there are perceptible and function al differences between the past and the present, at least for the ar ea that is owned by the county. The iconic ontological order conveys they message of ecological preservation, and any connection to the past must be bro ught into being through the discursive order. The same may be said for the urba n neighborhoods of Tampa, where operational order is not significantly di fferent despite changes in the configuration of the built environment, and the iconic order still conveys a sense of urbanization. However, there are still affect ive, nominal, and arbitrary

PAGE 134

126 distinctions that divide the neighborhoods of Tampa, a nd the bounds of those neighborhoods may be distinguished by residents through a symbolic message from associations with the built environment. With that said, it would be erroneous to claim that on ly one ontological order was involved with any of the settlements, or ind ividual structures within settlements. Moreover, it would be a mistake to claim that there was only one interpretant associated with a settlement or building. For example, within the symbolic order, a church may represent a spiritual connect ion with a deity and a human connection with distant communities united by th e ecumenical body of a particular denomination. Furthermore, to extend the example of the church, a single building may be perceived through the lens of sy mbolic (e.g. as a place of worship), indexical (e.g. doors as appropriate points of ingress and egress for congregants), and iconic interpretants (e.g. a cross-shape d floor plan). It seems reasonable to expect that residents of Bealsvill e will be the most knowledgeable about and the historical development, an d likely to provide immediate refinements to the results reported here. In urban Tampa, it may be necessary to present these results, along with the result s of previous research, to develop a core group of public stakeholders that will h ave knowledge to contribute to an improved understanding of the full t ime depth of those communities. For Willow, focusing on the site’s contribu tion to African American history in Hillsborough County may or may not be success ful in identifying a core group of public stakeholders that could provide refinem ent of our knowledge about those aspects of Willow. However, by escalating ou r search for such

PAGE 135

127 interested public stakeholders to a broader area, such as countywide or westcentral peninsular Florida, the descendants of Willow employees may still be able to contribute firsthand knowledge about life at that p articular company town. It may be more productive to research spatial data from c omparable company towns, and seek out interested public stakeholders on tha t front. One outcome of this research that has not been fully re alized is the value of the maps as tangible objects that can be taken back to the descendent communities and used as the basis for future work. Since there has almost certainly been information that was not recorded on ma ps, or in other primary documents, the maps can provided a starting point for re defining the uses of the represented space through time. In the settlements d iscussed here, there are many options available for bringing Communities of In terest into the sphere of participatory design of future research. It may also be prudent to include Communities of Interest that are drawn from stakeholde rs with interests other than heritage management, such as urban economists and community planners. In particular, the information regarding residential centralization, concentration, and clustering should be connected with planning for pu blic transportation, with special attention to any job centralization that has no t been covered here. The more typical approach of presenting research results to an audience that is brought together temporarily, to be told the results of a researcher’s particular program of research is not going to be entir ely eliminated; but, those lectures must be used as a basis for developing interest, e nrolling active participants in future research, and mobilizing those pe ople toward interaction,

PAGE 136

128 both between themselves and researchers and among themse lves. Given the availability of internet connections in these communit ies, whether on the campus of the Glover School or in the neighborhood public li braries of urban Tampa, a publicly viewable forum is a simple means of creating discu ssion space that can be implemented as open to all who are willing to cont ribute. As the research grows, collaborative collections management software can b e designed to allow participation, with varying levels of permission to uplo ad contributions that bring together multiple lines of research and public interest. This does not negate the importance of ethical collection of materials, or rigor ous analytical procedures; but, it does open up the processes of data collection an d analysis to a wider group of research partners. We do not need to give u p our position as experts in a particular style of research, or reduce the importan ce of scholarship. However, we will need to accept a role as facilitators of the de velopment of a common understanding between practitioners and interested nonspecialists from multiple communities.

PAGE 137

129 Chapter 8: Conclusion Although the two rural populations do appear to ha ve had other considerations in mind when placing the earliest house sit es, with agricultural potential being a greater concern than church building s as social focal points. However, as the communities grew, there do seem to have been some differences in preference for church locations. In par ticular, the Springhead community did appear to experiment with more edge-fo cused locations for their Baptist church; this would seem to relate to better acce ss to transportation networks leading to more contact with surrounding communi ties, such as Mulberry, Coronet, and Plant City. On the other ha nd, Bealsville’s Baptist church was moved back toward the center of the community, after much of the land on the southern boundary of the original community’s ext ent was lost to land speculation. Undoubtedly, the Baptist church is a focal point of the community, as indicated by their recent celebration of their 140 th anniversary, in November 2008; however, that may be attributable not only to being a social focal point of the community, but also a reflection of the isolation experienced by an historically black community surrounded by white communities. The iso lation of the community was not entirely undesirable during the oppr essive years of the Jim Crow South, for reasons of both safety and personal lib erty (Bethel 1981:95; Klugh 2004:114).

PAGE 138

130 The work that been done on urban Tampa, preliminary though it may be, should be used as a basis for further study of the effects of residential centralization on employment, and the suitability of mass transit in Hillsborough County. It should also provide a basis for identifyin g more areas of historic value within the city, and offer some potential areas for fu rther archaeological exploration. Due to urban renewal the built enviro nment has changed substantially, and continues to be altered. One of th e advantages of having a permanent GIS database of historical changes in the buil t environment is that archaeological exploration can be matched with planned construction. Tragically, successive episodes of construction may have been conducted i n an area before it was the value was realized; but, by keeping track of the spatial locations of those areas, we may be able to take advantage of the regular cycle of demolition and construction. The level of property division seen at Bealsville, alt hough less than seen at Springhead, is still consistent with what is reported for other historically black communities (Bethel 1981), and it does likely reflect limited property ownership opportunities, following the brief period of the Re construction. With fewer opportunities to acquire new lands, there was some migr ation away from the rural community of Bealsville, presumably toward urban centers such as Tampa; however, Bealsville is also remarkable in the sense that, in at least one case, descendents of the original GLO deed recipients still hol d part of the properties that were originally deeded to family members. So, in the case of Bealsville, the level of property division is probably just as attribu table to the pattern of land loss

PAGE 139

131 and reacquisition that community founders like Alfred Be al struggled to maintain. Future comparisons between communities from other areas, with other economic bases, will likely provide insight into environmental a nd economic reasons for differences in the way property becomes subdivided; of course, each new location will also have to account for socio-cultural diff erences that may have impinged on people’s ability to acquire new land, bey ond general availability of land. A GIS approach makes for a good foundation, but it is better at handling data that are stable over long periods of time, such a s long-term environmental data. However, it is also quite useful for representi ng both types of conjunctures, whether tracking patterns of population increases and decl ines, the material inertia of a settlement’s built environment, or the aggregation and division of land as property. GIS can be employed in representing ev ents, but it is far more suited to singular, major events than it is to the mund ane events that that occur in daily life. The representation of overlapping data is highly dependant on available data and methods used to collect that data. We can see the remnant path of church moved from th e perimeter of a town to the center, but we cannot distinguish where the mule that pulled the building rested. We can see the linear ditches that ha ve served to drain excess water from agricultural fields, perhaps even periods wh en the ditches were allowed to fill with sediment, but we cannot represent Andrew William’s grubbing hoe or trace the order in which he cleared roads and di tches. Through the course of collecting data, it has been the remembrances of people that provided

PAGE 140

132 a sense of boundary and motion. In some cases, those re membrances were written down; but, in other cases, it has been chance e ncounters with living people that provided insight into how people perceive d and used their built environment. As examples, when Bealsville residents knew that this study was also considering life in urban Tampa, they would menti on the Garrison neighborhood or recount a memory of riding the train into Tampa to attend school at higher grades than were available at the Glover S chool. Thus, what is presented here is a good understanding som e of the nodes in a county-wide network, and a basis for continuing re search on how people moved through those nodes. The data that has been col lected and stored through GIS can form the basis for filling in informat ion about the physical world between the dates that have been used here because ther e was comparable information for all settlements. However, the result ing synchronic maps can also be used to elicit more personal recollections from inter ested stakeholders. Those remembrances may be used as oral histories, but a creative use of GIS might also allow researchers to create data layers that reflect the semiosphere of individual informants. Each layer might represent on e person’s perceptions, during one particular time period, but they may also be used comparatively to better understand how people perceived their world di fferently. We must still overcome the limitations of how to represent brief ev ents, but that still leaves perceptions of long-term trends. However, that makes GI S a viable tool for reflecting not only physical space, but also cognitive spa ce.

PAGE 141

133 References Cited Aijmer, Goran 2000 Introduction: The Idiom of Violence in Imagery a nd Discourse In Meanings of Violence: A Cross Cultural Perspective edited by G. Aijmer and J. Abbink, pp. 1-22. Berg, New York. 2003 New Year Celebrations in Central China in Late Imper ial Times The Chinese University Press, Sha Tin, N.T., Hong Kong. Aiken, Charles S. 1985 New Settlement Pattern of Rural Blacks in the Ame rican South. Geographical Review 75(4):383-404. Ashmore, Wendy 2002 "Decisions and Dispositions": Socializing Spatial Arch aeology. American Anthropologist 104(4):1172-1183. Bailey, Geoff 2008 Time Perspectivism: Origins and Consequences In Time in Archaeology: Time Perspectivism Revisited edited by S. Holdaway and L. Wandsnider, pp. 13-30. The University of Utah Press, Glen Canyon. Bailey, Ron 1973 Economic Aspects of the Black Internal Economy. The Review of Black Political Economy 3(4):43-72. Barker, Eirlys M. 1984 Seasons of Pestilence: Tampa and Yellow Fever, 1824-1905 Unpublished Master's thesis, Department of History, Univ ersity of South Florida, Tampa. Bethel, Elizabeth R. 1981 Promiseland, a Century of Life in a Negro Community University of South Carolina Press, Columbia. Binford, Lewis R. 1981 Behavioral Archaeology and The" Pompeii Premise" Journal of Anthropological Research 37(3):195-208.

PAGE 142

134 Bintliff, John 2004 Time, Structure, and Agency: The Annales, Emergen t Complexity, and Archaeology In Blackwell Companion to Archaeology edited by J. Bintliff, pp. 174-194. Blackwell Publishing, Ltd., Mal den. Blanton, Richard E. 1994 Houses and Households: A Comparative Study Plenum Pub Corp., New York. Bloch, Marc 1992 The Historian's Craft Manchester University Press, Manchester. Boustan, Leah P. and Robert A. Margo 2009 Job Decentralization and Residential Location. Brookings-Wharton Papers on Urban Affairs :1-31. Brady, Rowena Ferrell and Canter Brown 1997 Things Remembered : An Album of African Americans in Ta mpa University of Tampa Press, Tampa. Braudel, Fernand 1995 The Mediterranean and the Mediterranean World in th e Age of Phillip Ii, Vol. 2 Translated by S. Reynolds. University of Califonia Press, Berkeley. Brown, Audrey L. 1994 Afro-Baptist Women's Church and Family Roles: Tran smitting Afrocentric Cultural Values. Anthropological Quarterly 67(4):173-186. Brown, Canter 1997 African Americans on the Tampa Bay Frontier Tampa Bay History Center, Tampa. 1998 Florida's Black Public Officials, 1867-1924 University of Alabama Press, Tuscaloosa. 1999 Tampa Before the Civil War University of Tampa Press, Tampa. 2000 Tampa in Civil War and Reconstruction University of Tampa Press, Tampa. Brown, Canter and Barbara Gray Brown 2003 Family Records of the African American Pioneers of Tampa and Hillsborough County University of Tampa Press, Tampa.

PAGE 143

135 Bryce, Herrington, Ernest Erber and Phillip L. Clay 1977 Black Urban Settlement Patterns: Trends and Prospect s. Habitat International 2(1-2):157-172. Burger, Oskar and Lawrence C. Todd 2006 Grain, Extent, and Intensity: The Components of Scale in Archaeological Survey In Confronting Scale in Archaeology: Issues of Theory and Practice edited by G. Lock and B. L. Molyneaux. Springer, New York. Burton, O. V. 2002 Reaping What We Sow: Community and Rural Histor y. Agricultural History 76(4):631-638. Campbell, Kenneth M. 1984 Geology of Hillsborough County, Florida [Electronic Reso urce] Open File Report (Florida Geological Survey) ; No. 6 Florida Geological Survey; PALMM, Tallahassee. Chang, K. C. 1967 Rethinking Archaeology Random House, New York. 1972 Settlement Patterns in Archaeology Addison-Wesley Modules in Anthropology, No. 24. Addison-Wesley Publishing Co., R eading. Dant, Tim 1999 Material Culture in the Social World : Values, Activiti es, Lifestyles Open University Press, Philadelphia. Deetz, James 1988 History and Archaeological Theory: Walter Taylor Revisited. American Antiquity 53(1):13-22. Drobney, Jeffrey A. 1996 Company Towns and Social Transformation in the N orth Florida Timber Industry, 1880-1930 The Florida Historical Quarterly 75(2):121145. Duncan, Pamela A n.d. The Willow Mill: An Oral History. Copy on file Robbins Manufacturing Co., Tampa. Dunn, Hampton 1977 Yesterday's Tampa E.A. Seemann Publishing Inc., Miami.

PAGE 144

136 Feinman, Gary M. 1997 Thoughts on New Approaches to Combining the Archae ological and Historical Records. Journal of Archaeological Method and Theory 4(3):367-377. Fischer, Gerhard 2000 Symmetry of Ignorance, Social Creativity, and M eta-Design. Knowledge-Based Systems 13(7-8):527-537. 2004 Social Creativity: Turning Barriers into Opportu nities for Collaborative Design Paper presented at the Participatory Design Conference, Toronto. 2007 Meta-Design: Expanding Boundaries and Redistribut ing Control in Design Paper presented at the 11th IFIP TC 13 internationa l conference on Human-computer interaction, Rio de Janeir o. Fletcher, Roland J. 1977 Settlement Studies (Micro and Semi-Micro) In Spatial Archaeology edited by D. L. Clarke, pp. 47-162. Academic Press, Lond on. 1984 Identifying Spatial Disorder: A Case Study of a Mongol Fort In Intrasite Spatial Analysis in Archaeology edited by H. J. Hietala and P. A. Larson, pp. 196-223. Cambridge University Press, New York. 1989 Social Theory and Archaeology: Diversity, Parado x and Potential. Mankind 19(1):65-75. 1991 The Messages of Material Behaviour: A Preliminary Discussion of Non-Verbal Meaning In The Meanings of Things: Material Culture and Symbolic Expression edited by I. Hodder, pp. 34-40. Routledge, New York. 1995 The Limits of Settlement Growth : A Theoretical Outli ne New Studies in Archaeology. Cambridge University Press, New Y ork. 2004 Materiality, Space, Time and Outcome In The Blackwell Companion to Archaeology edited by J. L. Bintliff, pp. 110-140. Blackwell Publishing Ltd., Malden. Florida Department of Environmental Protection 2007 LABINS Land Boundary Information System. Elect ronic document,, accessed March 15, 2007.

PAGE 145

137 Florida Department of Environmental Protection, Divi sion of State Lands, Board of Trustees Land Document System 2007 Land Document Search [GLO Sales Book Records]. Elect ronic document,, accessed March 15, 2007. Gates, Paul W. 1936 The Homestead Law in and Incongruous Land System. The American Historical Review 41(4):652-681. 1940 Federal Land Policy in the South 1866-1888. The Journal of Southern History 6(3):303-330. Glaeser, Edward L. and Jacob L. Vigdor 2001 Racial Segregation in the 2000 Census: Promising News The Brookings Institution, Center on Urban & Metropolitan Policy, Washington D.C. Glassie, Henry H. 1975 Folk Housing in Middle Virginia: A Structural Analysis of Historic Artifacts University of Tennessee Press, Knoxville. Goldfield, David R. 1991 Black Life in Old South Cities In Before Freedom Came: AfricanAmerican Life in the Antebellum South edited by E. D. C. Campbell, pp. 123-153. Museum of the Confederacy;University Press of Virginia, Richmond. Gotham, Kevin F. 2000 Urban Space, Restrictive Covenants and the Origins of Racial Residential Segregation in a Us City, 1900–50. International Journal of Urban and Regional Research 24(3):616-633. Greenbaum, Susan D. 2002 More Than Black: Afro-Cubans in Tampa University Press of Florida, Gainesville. Groves, Paul A. and Edward K. Muller 1975 The Evolution of Black Residential Areas in Late N ineteenthCentury Cities. Journal of Historical Geography 1(2):169-191. Gudeman, Stephen 2005 Community and Economy: Economy’s Base In A Handbook of Economic Anthropology edited by J. G. Carrier, pp. 94-106. Edward Elgar Publishing, Inc., Northampton.

PAGE 146

138 Hammer, yvind, David A. T. Harper and Paul D. Ryan 2001 PAST: Paleontological Statistics Software Package f or Education and Data Analysis. Palaeontologia Electronica 4(1):1-9. Harris, Trevor M. 2006 Scale as Artifact: GIS, Ecological Fallacy, and Arch aeological Analysis In Confronting Scale in Archaeology: Issues of Theory and Practice edited by G. Lock and B. L. Molyneaux, pp. 39-53. S pringer, New York. Hewitt, Nancy 2001 Southern Discomfort: Women’s Activism in Tampa, Flo rida. University of Illinois Press, Chicago. Hilliard, Sam B. 1994 Plantations and the Molding of the Southern Lan dscape In The Making of the American Landscape edited by M. P. Conzen, pp. 104126. Routledge, New York. Hillsborough County, Department of Planning and Zoni ng 1941 Preliminary Report Land Use Planning, Hillsborough Cou nty, Florida Hillsborough County Planning and Zoning Dept. Hillsborough County, Planning & Growth Management 1998 Hillsborough County Historic Resources Survey Report Submitted to Florida Department of State, Division of Historic R esources, Tallahassee. Hillsborough County and City of Tampa Water Atlas 2007 Data and Mapping. Electronic document,, accessed June 20, 2007. Howard, Walter T. and Virginia M. Howard 1994 Family, Religion, and Education: A Profile of A frican-American Life in Tampa, Florida, 1900-1930. Journal of Negro History 79(1):1-17. Howe, Kathleen S. 1997 Stepping into Freedom : An Analysis of the African-Amer ican Community in Hillsborough County, Florida During the R econstruction Era. Unpublished Master's thesis, Department of History, Univ ersity of South Florida, Tampa. 1999 Stepping into Freedom: African Americans in Hillsb orough County, Florida, During the Reconstruction. Tampa Bay History 20(2):4-30.

PAGE 147

139 Johnson, Matthew 2010 Archaeological Theory: An Introduction Blackwell Publishing, Ltd., Malden. Joseph, J. W. 2000 Archaeology and the African-American Experience in the Urban South In Archaeology of Southern Urban Landscapes edited by A. L. Young, pp. 109-126. University of Alabama Press, Tuscalo osa. Kellogg, J. 1977 Negro Urban Clusters in the Postbellum South. Geographical Review 67(3):310-321. Kerstein, Robert 1998 From Annexation to Urban Renewal: Urban Develo pment in Tampa During the 1950s and 1960s. Tampa Bay History 19(1):69-92. King, Thomas F. 2003 Places That Count: Traditional Cultural Properties in Cultural Resource Management AltaMira Press, Walnut Creek. Kinser, Samuel 1981 Annaliste Paradigm? The Geohistorical Structuralism of Fernand Braudel. The American Historical Review 86(1):63. Klejn, Leo S. 2001 Metaarchaeology. Acta Archaeologica 72(1):1-149. Klugh, Elgin L. 2004 African American Schoolhouses : Community, History, and Reclamation. Unpublished Ph.D. dissertation, Department of Anthropology, University of South Florida, Tampa. Kowalewski, Stephen A. 1997 A Spatial Method for Integrating Data of Diffe rent Types. Journal of Archaeological Method and Theory 4(3):287-306. LaMotta, Vincent M. and Michael B. Schiffer 2001 Behavioral Archaeology: Toward a New Synthesis In Archaeological Theory Today edited by I. Hodder, pp. 14–64. Blackwell Publishers, Inc., Malden.

PAGE 148

140 Landry, Shawn Michael, Naomi Oliver and Cheran Willi ams 2004 A Guide to Historic Architectural Resources and Historic Preservation of Unincorporated Hillsborough County University of South Florida, Florida Center for Community Design a nd Research, Tampa. Latour, Bruno 2005 Reassembling the Social. An Introduction to Actor-Networ k-Theory Oxford University Press, New York. Lee, David 1992 Black Districts in Southeastern Florida. Geographical Review 82(4):375-387. Lock, Gary and Brian L. Molyneaux 2006 Introduction: Confronting Scale In Confronting Scale in Archaeology: Issues of Theory and Practice edited by G. Lock and B. L. Molyneaux. Springer, New York. Lotman, Juri M. 2005 On the Semiosphere. Sign Systems Studies 33(1):205-226. Marler, Scott P. 2004 Fables of the Reconstruction: Reconstruction of the Fables. The Journal of The Historical Society 4(1):113. Marquardt, William H. 1992 Dialectical Archaeology In Archaeological Method and Theory, Volume 4 edited by M. B. Schiffer, pp. 101-140. University of Arizona Press, Tucson. Massey, Douglas S. 2001 Residential Segregation and Neighborhood Condit ions in U.S. Metropolitan Areas In America Becoming: Racial Trends and Their Consequences, Volume 1 edited by N. J. Smelser, W. J. Wilson and F. Mitchell, pp. 391-434. National Academy Press, Washingt on, D.C. Mitchell, Thomas W. 2000 From Reconstruction to Deconstruction: Undermining B lack Landownership, Political Independence, and Community t hrough Partition Sales of Tenancies in Common. Nw. UL Rev. 95:505. Nicholls, William H. 1964 The South as a Developing Area. The Journal of Politics 26(1):2240.

PAGE 149

141 Osman, Zehra 2002 Saving a Few, before Losing Them All-a Strategy for Setting Priorities. CRM 25(3):9-13. Panamerican Consultants 2001 An Archaeological and Historical Survey of the Perry Harvey Senior Park, Downtown Tampa, Hillsborough County, Florida. Report prepared for Award Engineering, North Florida Avenu e, Tampa. Parker, Patricia L. and Thomas F. King 1998 Guidelines for Evaluating and Documenting Trad itional Cultural Properties. U.S. Department of the Interior, Nationa l Park Service, National Register of Historic Places, Washington D.C. Parker, William N. 1980 The South in the National Economy, 1865-1970. Southern Economic Journal 46(4):1019-1048. Parsons, Jeffrey R. 1972 Archaeological Settlement Patterns. Annual Reviews in Anthropology 1(1):127-150. Pizzo, Anthony P. 1969 Tampa Town, 1824-1886: The Cracker Village with a Lat in Accent Hurricane House Publishers, Coconut Grove. Plog, Fred T. 1973 Diachronic Anthropology In Research and Theory in Current Archaeology edited by C. L. Redman, pp. 181-198. John Wiley an d Sons, New York. Pope, Christie F. 1970 Southern Homesteads for Negroes. Agricultural History 44(2):201212. Preucel, Robert W. and Alexander A. Bauer 2001 Archaeological Pragmatics. Norwegian archaeological review 34(2):85-96. Publication of Archival Library and Museum Materials 2007 World Map Collection. Electronic document,, accessed March 21, 2007.

PAGE 150

142 Quantum GIS Development Team 2010 Quantum GIS Geographic Information System. Open Source Geospatial Foundation Project. Raper, Arthur Franklin, J. H. McGrew and Benjamin E. M ays 1927 A Study of Negro Life in Tampa : Made at the Request of the Tampa Welfare League, the Tampa Urban League and th e Tampa Young Men's Christian Association s.n., Tampa. Rapoport, Amos 1982 The Meaning of the Built Environment: A Nonverbal Communication Approach University of Arizona Press, Tucson. 1990 Systems of Activities and Systems of Settings In Domestic Architecture and the Use of Space: An Interdisciplinary C ross-Cultural Study edited by S. Kent, pp. 9-20. Cambridge University Press, New York. Rivers, Larry E. 2000 Slavery in Florida: Territorial Days to Emancipation University Press of Florida, Gainesville. Roback, Jennifer 1984 Southern Labor Law in the Jim Crow Era: Exploit ative or Competitive. U. Chi. L. Rev. 51:1161. Robb, John E. 1998 The Archaeology of Symbols. Annual Review of Anthropology 27:329-346. Robinson, Ernest L. 1928 History of Hillsborough County, Florida The Record Company, St. Augustine. Shelley, Fred M. and Gerald R. Webster 1998 Population, Settlement, Race, and Ethnicity in t he South. Journal of Geography 97(4):163-175. Silver, Christopher 1997 The Racial Origins of Zoning in American Cities In Urban Planning and the African-American Community: In the Shadows edited by J. M. Thomas and M. Ritzdorf, pp. 23-42. Sage Publications, Inc., Thousand Oaks.

PAGE 151

143 Smith, Eugene A. 1884 Report on the Cotton Production of the State of Florida with an Account of the General Agricultural Features of the Sta te In Census Reports Tenth Census. June 1, 1880 edited by F. A. Walker, C. W. Seaton and E. W. Hilgard, pp. 175-257. Department o f the Interior, Census Bureau, Washington D.C. Smith, Michael E. 1992 Rhythms of Change in Postclassic Central Mexico: Archa eology, Ethnohistory, and He Braudelian Model In Archaeology,Annales, and Ethnohistory edited by A. B. Knapp, pp. 23-34. Cambridge Unive rsity Press, New York. Tampa-Hillsborough County Public Library System 2007 Digital Sanborn Maps. Electronic document,, accessed Ma rch 23, 2007 Taylor, Walter W. 1983 A Study of Archaeology Southern Illinois University at Carbondale, Center for Archaeological Investigations, Carbondale. Thompson, Raymond H 2000 Edgar Lee Hewett and the Political Process. Journal of the Southwest 42(2):271-318. Trigger, Bruce 1967 Settlement Archaeology: Its Goals and Promise. American Antiquity 32(2):149-160. Tubbs, Sharon 2008 Marker Pays Homage to Dobyville, a Caring Villag e In St. Petersburg Times : 28 March: TCT1, TCT7. St. Petersburg. United States Census Bureau 1958 Population Schedules of the Tenth Census of the U nited States, 1880 (Florida). [Microform] In National Archives microfilm publications ; T623, reels 125-132. edited by U.S. Bureau of the Census. National Archives, Washington, D.C. 1965 Population Schedules of the Ninth Census of the U nited States, 1870, Florida [Microform], edited by U.S. Bureau of the Census. National Archives, Washington, D.C.

PAGE 152

144 1978 12th Census of Population, 1900, Florida [Microf orm] In National Archives microfilm publications ; T623, reels 165-177. edited by U.S. Bureau of the Census. United States, National Archives and Records Service, Washington, D.C. University of Alabama 2007 Historical Map Archive. Electronic document,, a ccessed June 20, 2007. Wacquant, Loic J. D. 1997 Three Pernicious Premises in the Study of the Amer ican Ghetto. International Journal of Urban and Regional Research 21(2):341-353. Weinberg, Bruce A. 2000 Black Residential Centralization and the Spatial Mismatch Hypothesis. Journal of Urban Economics 48:110-134. Weisman, Brent R. and Lori D. Collins 2004a A GIS Archaeological Model and Testing of Nine Elapp Preserves, Hillsborough County, Fl Submitted to Hillsborough County ELAPP. 2004b A Model for Evaluating Archaeological Site Significanc e in Cities: A Case Study from Tampa, Florida Submitted to Florida Department of Transportation. White, William A. 1970 The Geomorphology of the Florida Peninsula Bureau of Geology, Division of Interior Resources, Florida Dept. of Natura l Resources Tallahassee. Willey, Gordon R. 1953 Prehistoric Settlement Patterns in the Viru Valley, Per u US Govt. Print. Off., Washington, DC. 1989 Settlement Pattern Studies and Evidence for Int ensive Agriculture in the Maya Lowlands In Archaeological Thought in America edited by C. C. Lamberg-Karlovsky, pp. 167-182. Cambridge Univ ersity Press, New York. Woodson, Robert L. 1989 Race and Economic Opportunity. Vand. L. Rev. 42:1017.

PAGE 153

145 Wright, Gavin 1987 The Economic Revolution in the American South. The Journal of Economic Perspectives 1(1):161-178.