The People of the Eye: Deaf Ethnicity and Ancestry (34 page)

RECESSIVE TRANSMISSION

Hearing parents

In the pedigrees of Figures 2 through 17, we examined the children of thirty-seven hearing couples who had at least one Deaf child and found 110 Deaf and 172 hearing yielding a total of 282 children. With recessive transmission, we would expect by chance at most 25 percent of the total number of children to be Deaf, that is, 71 children. However, 39 were. This discrepancy is statistically reliable. We may have underestimated the number of hearing children in the following ways. We went to great pains to identify all Deaf children and endeavored to accurately ascertain the numbers in their sibships. By beginning with the Deaf members of sibships, however, we risked overlooking consanguineously related hearing parents who had the necessary recessive genes but, by chance, had no Deaf children and whose hearing children, then, are not included in our figures. Furthermore, it is much easier to overlook a hearing child than Deaf children because several sources concur in identifying Deaf children (school registers, Fay's census of Deaf marriages, etc.) Such sources of bias are known as ascertainment bias.

Deaf parents

In the pedigrees of Figures 2 through 17, we examined the children of thirty-seven couples with both members Deaf. We found forty Deaf and eighty hearing yielding a total of 120 children. With recessive transmission, we would expect a Deaf couple to have all Deaf children (provided that the parents are both Deaf by virtue of the same genes.) Three couples among the thirty-seven are known to be consanguineous. Two of them had only Deaf children, as expected and the third had some hearing children. An additional two Deaf couples, with one member dominant and one recessive, were examples of dominant transmission; they had six Deaf and five hearing children, as expected.

The marriages of Freeman SmithD and Deidama WestD (Fig. 7, Lambert pedigree), and of Benjamin MayhewD and Hannah SmithD (Fig. 4, Mayhew pedigree) were consanguineous. Therefore, the members of each couple most likely had the same pair of recessive genes and all of their children are expected to be Deaf, which was the case. (The Deaf-Deaf marriage of Thomas BrownD and Mary SmithD is different because, as we have maintained, Thomas BrownD inherited a dominant gene from his father, NahumD.) Nevertheless, contrary to expectations, some of the Deaf-Deaf marriages also had hearing children. Franklin TiltonD and Sarah FosterD (Fig. 3, Tilton pedigree) had some hearing children as did Rebecca WestD and Eugene TraskD (Fig. 7, Lambert pedigree), Jacob BosworthD and Sally Allen (Fig. 9, Allen pedigree); and George WestD and Sabrina RogersD (Fig. 9, Allen pedigree). Unlike the consanguineous Deaf couples discussed above, husband and wife were not known to be related in any of these couples. When it comes to such Deaf-Deaf couples whose members are unrelated, we can make no prediction about the numbers of Deaf and hearing children they will have.

In some cases Deaf couples had only hearing children. The CurtisRowe pedigree (Fig. 17) shows three marriages between Deaf partners with all told six hearing and no Deaf children. If the parents shared the same genes, we would expect all the children to be Deaf. Setting aside the marriage of Benjamin RoweD and Ann CurtisD who had no children, it is likely that the parents of the six hearing children were Deaf because of different gene pairs. Looking at the marriage of Ebenezer CurtisD and Lucy RoweD, we see that Ebenezer'sD father, William B. Curtis, is descended from the Kentish progenitor by the same name. Hence EbenezerD may be Deaf owing to Kentish genes. His wife, Lucy RoweD, has no known ancestry in Kent which may explain why their two children are hearing, since a pair of Deaf parents with different genes will not have Deaf children. The situation for George CurtisD and Nancy RoweD is the same and their four children were all hearing.

SPECIAL_IMAGE-page0205_0000.svg-REPLACE_ME

SPECIAL_IMAGE-page0205_0001.svg-REPLACE_ME

PEDIGREE SOURCES

We began our inquiry into the early Deaf families of Henniker, New Hampshire, Martha's Vineyard, Massachusetts, and southern Maine by identifying Deaf individuals in those locales, using the 1850 census, which listed 266 "Deaf and Dumb" persons living in Maine. We retained only those individuals whose family names occurred twice or more.20 This principle includes in the set of hereditarily Deaf a few people whose surnames happen to coincide but are not related-such as different Brown families. We have endeavored to ferret those out. The principle excludes some singletons who are in fact hereditarily Deaf. Next we searched for ascendants, descendants, and siblings of those Deaf men and women who had been retained, using numerous general sources and Deaf-related sources. Among the general sources we include other censuses; beginning in 1830, the federal census reserved a column for "Deaf and Dumb" but only a count was given next to the household-individuals were not identified until the 1850 and later censuses. We also used town histories, vital records, biography, and genealogy accessed in hard copy or through the internet. Our primary internet services were FamilySearch.org and Ancestry.com.

Specific source references are given in the notes associated with the corresponding family and town. We also made extensive use over the years of the New England Historic Genealogical Society, which has superb resources, both human and documentary, including many pertinent manuscripts. Likewise, the staff and collections of the Maine Historical Society in Portland and the Maine State Archives in Augusta were very helpful.

Among Deaf-related sources we found particularly valuable the rolls of the American School for the Deaf from its opening to May 1, 1887, published in the school's 71st Annual Report, and a copy of school registry entries for students from Maine, provided by the school archivist, Gary Wait.21 In 1817, Massachusetts conducted a census of Massachusetts and Maine school-age Deaf children associated with paying their tuition at the American Asylum.22 We also made use of state finance records of payments made to the American School, and of attendance lists from the four reunions of the American School in 1850, 1854, 1860, and 1866. The Gallaudet University Archives provides a free online facility to search some forty Deaf publications as well as Gallaudet alumni association obituaries and records of the National Fraternal Society of the Deaf.

The membership rolls of the New England Gallaudet Association of Deaf-Mutes and of the Maine Deaf-Mute Mission were also valuable. The notebooks of Alexander Graham Bell, kindly photocopied in part at our request by the Volta Bureau, provided a rich lode of Deaf names. It is not surprising that a century after those notes were written, and with the tools now available to us, we find errors in identifying Deaf people and their ancestry. Nevertheless, the voluminous notebooks bear witness to Bell's deep and abiding interest in Deaf ancestry. Bell also provided information on Deaf families to a Royal Commission and to a journal concerned with inheritance.23

Finally, among Deaf-related sources, and of great importance, we cite the survey of Deaf marriages conducted by Edward Allen Fay.24 In 1898, Fay, a professor at Gallaudet College (now Gallaudet University), published a nationwide sample of pedigrees on 30,000 individuals in 4,471 marriages between 1803 and 1894 in which one or both partners were Deaf. Marriages between two Deaf people accounted for 76 percent of the marriages.25 The information in the Fay book was obtained from Deaf marriage partners and, occasionally, from Deaf educators or others who knew them. The data included the birth dates of the marriage partners, the number of Deaf and hearing siblings that each possessed, age at becoming Deaf and assigned cause, marriage date, schools attended, numbers of children Deaf and hearing, crossreferences to Deaf relatives, and helpful remarks (such as identifying a spouse as hearing).

Family names have been replaced by reference numbers in the Fay book, so two files supplied by the Gallaudet University Archives are indispensable: Fay Index Husbands and Fay Index Wives. The Fay book is actually a condensation of data on survey forms, which can be found at the Gallaudet University Archives, the Volta Bureau, and (with a subscription) on the internet at Ancestry.com. These forms contain a wealth of additional material, such as the names of the husband and wife's parents.

CAUTIONS ON SOURCES

We could not have pursued this research without access to genealogic information through the internet. Genealogies are usually constructed for one family. The task of creating pedigrees for a large group of families was out of reach for most genealogists until the recent advent of internet services. Using the internet comes at a price-loss of accuracy. It is not uncommon for two equally good sources to give conflicting information about dates, locales, and even ancestors. Fortunately, there are many constraints on descendants and their dates. For example, an individual's birth date must fall within a given range to conform to those of his or her parents and children. Furthermore, we complemented research on the internet with the resources of two excellent genealogical libraries: the New England Historic Genealogical Society and the Maine Historical Society.

In the nineteenth century, the chances of an infant surviving to age five were one in three, so many Deaf and hearing children could be missed in the decennial censuses. The wives' maiden names were not given. The census sheets were sometimes inconsistent, identifying persons as Deaf in one census and failing to identify them so in the next. The spelling of names was at times faulty and inconsistent. Most censuses report the age, not the date of birth, of persons enumerated. We subtract the age that was reported from the date of the census to obtain the invariant year of birth; however, that is accurate only plus or minus one year.

In early vital records people are occasionally said to be "of X" whereas in fact they moved to X. For example, a Mayflower passenger is said to be of London, when his or her birthplace was Kent. Conflicting information is occasionally found when comparing sources, so pedigree assignments are made on the weight of the evidence, knowing that some will prove erroneous. The numbers of hearing and Deaf offspring may be inaccurate because of the high perinatal mortality of the time. It is a commonplace with genealogic information on the internet that parents are listed with only a subset of their children; thus the number of hearing siblings must be used with great caution when deciding whether trait transmissions were dominant or recessive. The form of the gene associated with the Deaf trait may be variably expressed-that is, some individuals may be hard of hearing and underreported as Deaf in some censuses; our pedigrees for hereditarily Deaf persons, do not distinguish between hard of hearing and Deaf. The hearing status of early ancestors is difficult to ascertain accurately. In some cases paternity is also in doubt.

FORMING PEDIGREES

Our focus has been on people who were direct ancestors of Deaf people. Therefore, analyses that require a complete enumeration of the families of the ancestors cannot be accomplished with our data. We did not include in our quantitative analyses families for which we were not confident of the sibship, the numbers of Deaf and hearing children. We generally did not extend the search for Deaf descendants beyond 1900 as our focus was on early Deaf families. Further, we wished to respect the privacy of living individuals. However, all our information came from public sources or from family members. Once we undertook to discover the genealogy of a Deaf family, our search for Deaf members sometimes extended over state boundaries or into the twentieth century. We included out-of-state Deaf individuals when it seemed helpful but many of the pedigrees would contain more Deaf members if a regional search were conducted.

The pedigrees presented in this book also appear on the web along with many others: http://dvn.iq.harvard.edu/dvn/dv/DEA. For legibility in this book, pedigrees have been reduced by pruning members unrelated to the Deaf and by purging all information except first and last name and dates of birth and death. These same pedigrees appear at the website in emf and pdf format, also pruned but with full information on each individual restored. Finally, numerous pedigrees are presented, pruned and with full information, that could not be accommodated in this book. "Pruning" involves two stages. In the first, the pedigree includes all the candidates for ancestor, child, or sibling of a Deaf person; this first stage is as inclusive as possible. Then, in the second stage, the inclusive group was pruned retaining only the Deaf people, their ancestors, their descendants, and their siblings-no one else. (We thank Jason Freitas for his masterful programming of data collection, analysis and reduction.) Diagramming of lineages was achieved with Cyrillic software.

Each pedigree gives the descendants of the named progenitor who are in the line of descent to a Deaf person, as well as that Deaf person's siblings and descendants. Readers looking into genealogy should check each of the multiple entries for a given individual in the Every Name Index for the pedigrees on the website (Appendix D). The website "workbook" contains many partial pedigrees, where diligent effort did not yield solutions. Despite its incompleteness, we have put our workbook on the web because it contains pedigree information for many more families than this book can accommodate and because we wish to assist those who are interested in studying the genealogies and family histories. In that regard, we welcome corrections and additions whose sources are substantiated.

Genealogies usually trace ascendants only as far as the first male of the given family name who immigrated to America; that person is called the "progenitor." The pedigrees are organized around the male ancestor in part because children and their mothers take on the male ancestor's name. However, the hereditarily Deaf child may have received this trait from the paternal lineage, the maternal lineage, or both; we did not encounter sex-linked transmission of the Deaf trait. We stress that both lineages are equally important for genetic transmission but, in many families, maternal lineages were impossible to trace because the maiden name of the mother was not given or the documents to which we had access were organized to present only the male lines. We may, therefore, have overlooked some consanguineous relationships.