¹ Whereas an alphabet comprises a set of characters representing the individual sounds of a language, a syllabary comprises a set of characters representing a language's combined sounds of a consonant followed by a vowel (e.g. ba, be, bi, bo, bu). The syllabaries discussed in this paper, Cherokee and Vai, have basically two types of characters, most representing a consonant-vowel (CV) combination and the remainder representing an independent vowel (V). Syllabaries necessarily have more characters than alphabets and have generally been used for languages which, unlike English, allow almost exclusively CV or VCV syllables.

² Dalby, David, “A Survey of the Indigenous Scripts of Liberia and Sierra Leone: Vai, Mende, Loma, Kpelle and Bassa,” African Language Studies 8(1967), 10Google Scholar; see also idem., “The Indigenous Scripts of West Africa and Surinam: Their Inspiration and Design,” African Language Studies 9(1968), 167–68. Paul Hair had researched into Vai history from a teaching base in Sierra Leone.

³ See, for example, Dalby, David, “The Historical Problem of the Indigenous Scripts of West Africa and Surinam” in Dalby, David, ed., Language and History in Africa (London, 1970), 117Google Scholar; Gregersen, Edgar, Languages in Africa (New York, 1977), 188–89Google Scholar; idem., “Successes and Failures in the Modernization of Hausa Spelling” in Joshua Fishman, ed., Advances in the Creation and Revision of Writing Systems (The Hague, 1977), 423; Singler, John V., “Scripts of West Africa” in Daniels, Peter T. and Bright, William, eds., The World's Writing Systems (Oxford, 1996), 594.Google Scholar The possible connection was also referred to in a paper given at the 1972 Conference on Manding Studies at the School of Oriental and African Studies, University of London; the paper circulated and was later available on request, but never published: Stewart, Gail, “The Early Vai Script as Found in the Book of Ndole,” 19.Google Scholar

⁴ Holsoe, Svend E., “The Cassava-Leaf People: an Ethnohistorical Study of the Vai People with a Particular Emphasis on the Tewo Chiefdom” (PhD, Boston University, 1967), 51.Google Scholar Holsoe carried out field research in Liberia.

⁵ Note that the term “Liberia” as used in this paper refers to the area of the present-day state. Not recognized internationally until 1847, in its early years Liberia consisted of a few Afro-American settlements along the coast organized by several bodies, especially the American Colonization Society (hereafter ACS), with a main settlement at Monrovia, and the Maryland Colonization Society, with a settlement at Cape Palmas. Agents of the former body progressively operated along the coast northwest of Monrovia, supporting small trading settlements, and at times claiming their incorporation within “Liberia.” The chiefs at Cape Mount were willing to develop commercial relations with Monrovia, particularly after the British in the mid 1820s imposed an anti-slave trade blockade on the coast to the north (between Sierra Leone and Gallinas), which damaged Cape Mount trade in that direction; and the Afro-American settlers attempted to impose a monopoly on the trade to the south of Cape Mount (Gurley, Ralph Randolph, The Life of Jehudi Ashmun, late Colonial Agent in Liberia [New York, 1839], 319, 331Google Scholar). When Curtis settled at Cape Mount this district was not part of Liberia, but in 1830–31 he acted for the ACS in negotiations with chiefs at Cape Mount, which led to a treaty in December 1831 ceding the town of Bendu and its vicinity on condition that the Liberians establish a settlement and a school—neither condition being met, by 1833 the Vai regarded the treaty as void (Holsoe, , “Cassava-Leaf People,” 127–28Google Scholar). The acting Governor of Liberia, when he visited Cape Mount in early 1834, found that “the very Kings whose names are signed to the deed … deny the transaction and charge us with fraud” African Repository (hereafter AR) 10/2(04 1834), 55.Google Scholar However, one reason for the confusion was a local war between rival Vai polities and rulers, presumably ones with differing attitudes to anti-slave trade pressures and the Liberian encounter.

⁶ Holsoe, Svend E., “A Case of Stimulus Diffusion? (a Note on Possible Connections Between the Vai and Cherokee Scripts),” Language Sciences 15(04 1971), 22–24.Google Scholar

⁷ See, for example, Juleus, Nets, “What's an Indian Language doing in Africa?,” Science Digest (12 1973), 24–25Google Scholar; Singler, , “Scripts,” 594.Google Scholar

⁸ Goody, Jack, The Domestication of the Savage Mind (Cambridge, 1977), 18.Google Scholar

⁹ Gregersen, , Languages, 189.Google Scholar

¹⁰ Dalby, , “Historical Problem,” 119.Google Scholar

¹¹ Although not publishing on the Cherokee-Vai link, in the decades after the 1960s Paul Hair collected considerable material on both scripts, but working from a British base, in respect of the Cherokee script invention was unable to pursue the necessary archival research in the United States. Tuchscherer joined him in the research in 1996, after undertaking field research in Sierra Leone and the completion of a dissertation on West African scripts, and he undertook much of the later research represented in this paper, both in Britain and the United States.

¹² See, for example, the report of the Museum of the Alphabet at <www.jaars.org/museum/alphabet/galleries/africa.htm>.

¹³ A critique of Sequoyah's role in the invention of the Cherokee script will appear in a forthcoming article by the authors.

¹⁴ It would be difficult to identify a single locality where Sequoyah devised his script. He was born in Tuskegee, Tennessee, but as a young man moved to Willstown, Alabama, where he may have been already thinking about creating a script. Prior to the invention of his script, he had moved to Pope County, Arkansas, where it appears he finalized the work. He then returned to the east to introduce the script among his fellow Cherokees.

¹⁵ “Invention of the Cherokee Alphabet,” Missionary Herald (hereafter MH) 29/10(10 1828), 330Google Scholar; reprinted from the Cherokee Phoenix. Note that this account of Sequoyah and the invention of the Cherokee script differs in certain details from other accounts of the same episode.

¹⁶ Walker, Willard and Sarbaugh, James, “The Early History of the Cherokee Syllabary,” Ethnohistory 40(1993), 70–71.CrossRefGoogle Scholar

¹⁷ MH (October 1828), 331.

¹⁸ Walker, /Sarbaugh, , “Early History,” 71.Google Scholar

¹⁹ Bass, Althea, Cherokee Messenger (Norman, 1936), 33–34.Google Scholar

²⁰ Mooney, James, Myths of the Cherokees, Smithsonian Institution, Bureau of American Ethnology, 19th Annual Report, 1897–1898 (Washington D.C., 1900), 110.Google Scholar

²¹ Ibid., 110–11.

²² Ibid., 110.

²³ Walker, /Sarbaugh, , “Early History,” 71–72.Google Scholar

²⁴ Mooney, , Myths, 115.Google Scholar

²⁵ “Cherokees,” Cherokee Phoenix (1 01 1831)Google Scholar, as reproduced in Kilpatrick, Jack Frederick and Kilpatrick, Anna Gritts, New Echota Letters: Contributions of Samuel A. Worcester to the Cherokee Phoenix (Dallas, 1968), 87–88.Google Scholar Earlier that year, an act came into effect in the state of Georgia that denied the Cherokee basic civil rights, the right to own land, and the right to hold public meetings (Mooney, , Myths, 117).Google Scholar

²⁶ MH, 24/5(May 1828), 162.

²⁷ Worcester, Samuel, “Cherokee Alphabet,” Cherokee Phoenix (21 02 1828)Google Scholar, as reproduced in Kilpatrick, and Kilpatrick, , Letters, 7–9Google Scholar; this article also appeared in MH 24/5(May 1828), 162–63. Later in the same year, the editor of the Missionary Herald wrote that “probably no people in the world can learn to read their own language, when written, so easily as the Cherokees:” MH 24/10(October 1828), 331.

²⁸ “Report of Messrs. Wilson and Wynkoop,” MH 30/6 (06 1834), 215, 30/9 (September 1834), 336Google Scholar; reprinted in AR 10/6 (August 1834), 180–81.

²⁹ MH 30/6 (June 1834), 215.

³⁰ They decided on Cape Palmas, at the other end of the coast, as a suitable site for a mission station, and then returned to the United States, but Wilson brought a party to Cape Palmas to start a mission in December 1834: Tracy, Joseph, “History of the Missions of the American Board of Commissioners for Foreign Missions” in History of American Missions to the Heathen (hereafter HAM) (Worcester, MA, 1840), 252.Google Scholar

³¹ The journal entries and report were published in Bates, Oric, “The Origin of the Vai Syllabary,” Harvard African Studies 1(1917), 290–92.Google Scholar We reproduce the originals here because the Bates version contains a number of transcription errors, along with omissions of journal entry dates.

³² Harvard University, ABCFM archive (hereafter HU/ABCFM), “Rev. J. Leighton Wilson's Journal of his Visit to Western Africa; January 22, 1834 to February 28, 1834, Part I” in section Western Africa / “West Africa, South Africa. Previous to 1838” (53), ABC.15.1, vol. 1. Note that Wilson and Wynkoop were informed about the Vai script by the Rev. John Pinney, a Presbyterian missionary and at the time chief agent for the ACS settlement of Afro-Americans at Cape Mesurado; for how Pinney acquired knowledge of the script see note 34 below.

³³ HU/ABCFM, “Journal of S.R. Wynkoop to Western Africa, 1834,” in section Western Africa / “West Africa, South Africa. Previous to 1838” (53), ABC.15.1, vol. 1.

³⁴ At the end of January 1834, it is “not more than a year” since the script was “commenced”; in early March, it was commenced “about a year since;” in mid April, it is “not more than two years since it was first invented.” The first two references point to the script being “commenced” (whatever that precisely means) in early 1833, the third may indicate instead 1832. All three references seem to derive from one informant, Pinney, and they may be merely repeating his estimate of the date of invention. Pinney first arrived at Monrovia on 16 February 1833, and on 28 February was preparing to visit parts of the coast, probably after his period of “seasoning,” but he had returned to the United States by July. Although he “penetrated the interior as far as the natives would permit,” it is not certain that he visited Cape Mount and perhaps unlikely that he traveled as far, although he “acquired authentic information concerning other parts of the country [than the Monrovia district]”: AR 9/2(1833), 60–61, 94; HAM, 712. He returned as acting Governor of Liberia, arriving on 31 December 1833, and “visited Cape Mount soon after my arrival”: AR 10/2(April, 1834), 55. If Pinney learned about the invention when visiting Cape Mount, this is more likely to have been in 1834 than in 1833, and if so he must have visited during January, in order to inform Wilson and Wynkoop by the 30th of that month. If this was the case, and even if it were his words that were being repeated in the first two references, invention in early 1833 would seem correct. However, if in fact he visited in 1833, invention in 1832 would seem fairly certain. But another possibility remains, that he “acquired authentic information” from someone else, an unknown individual, either during 1833 or in early 1834, and simply repeated it. It is perhaps significant that no direct comment on the invention from him is known. All in all, given the uncertainty, it seems safer to suppose the invention to have begun in either 1832 or 1833.

³⁵ Various attempts to explain the shapes of the characters in terms of copying from earlier systems or as largely pictograms have been unrewarding, some being extremely far-fetched. (But perhaps less so than the claim that Cherokee borrowed from Vai, via a pre-Columbus Mande invasion of America: Winters, Clyde, “The Influence of the Mande Scripts on American Writing Systems,” BIFAN 39B (1977), 405–31.Google Scholar) For a scholarly critique which allows only slight traces of borrowing from the Roman and Arabic scripts, see Dalby, “Inspiration and Design.” Wynkoop's “merely fanciful,” that is independently imaginative, has merit, in our view.

³⁶ HU/ABCFM, “Report of Messrs. Wilson and Wynkoop of their Visit to Western Africa” in section Western Africa / “West Africa, South Africa. Previous to 1838” (53), ABC. 15.1, vol. 1. Wilson added: “I inquired particularly about the new system of writing and found the account I had correct. I saw one book in it and I procured a small specimen of the writing.” The specimen of the syllabary sent to the ABCFM mission rooms in Boston (reported also in the Missionary Herald and African Repository) was believed to have been lost, as there was no record of its ever having been displayed or even received, and searches by scholars in the ABCFM archives had failed to turn it up. In 1944 the ABCFM archives was deposited by the United Church Board for World Ministries at Harvard University, where we recently traced the specimen (see Figure 2). This appears to be the earliest extant manuscript in an indigenous script from sub-Saharan Africa. A transliteration and translation of this Vai manuscript by Tuchscherer is forthcoming. The author of the manuscript (not identified by Wilson) was one Fan Dawo Kelondo.

³⁷ Forbes, Frederick E., “Despatch Communicating the Discovery of a Native Written Character at Bohmar, on the Western Coast of Africa, near Liberia, Accompanied by a Vocabulary of the Vahie or Vei Tongue. Communicated by the Admiralty. Received April 23 1849,” Journal of the Royal Geographical Society of London 20(1851), 89–113.CrossRefGoogle Scholar In January 1849, when on his way to England, Forbes had informed the missionaries at Freetown of the existence of the script, and both parties passed the news to London, while almost immediately Koelle journeyed from Freetown to Cape Mount to investigate the script: Hair, P.E.H., “Notes on the Discovery of the Vai Script, With a Bibliography,” Sierra Leone Language Review 2(1963), 36.Google Scholar Forbes began his report: “It has fallen to my lot to make a discovery of such importance to the civilization of Africa that …” The highly-charged tone of this report by a layman resembles that of the missionaries who announced the Cherokee script, as it also resembles that of Wilson, who in 1858 reminiscing about the Vai script, described its invention as “one of the most remarkable achievements of this or any age:” Wilson, J. Leighton, Western Africa: its History, Condition, and Prospects (New York, 1856), 95.Google Scholar Similarly, in 1849 the preface written on behalf of the CMS (the writer anonymous, but probably Henry Venn, the Secretary of CMS) in a printed pamphlet containing Koelle's report began in terms of the contemporary moral correctness with the following remarks. “It has been frequently asserted that no attempt has been made by the Native Tribes of the Western Coast of Africa to reduce their languages to writing. … This fact has been often alleged as a proof of the low intellectual qualities of the natives. … But… Mr. Koelle's Narrative affords such a proof of intellectual ability and enterprize in the Natives, as well as of a certain degree of moral and religious feeling—that it holds out bright hopes of the introduction of civilization and Christianity …” Koelle, S.W., Narrative of an Expedition into the Vy Country of West Africa and the Discovery of a System of Syllabic Writing Recently Invented by the Natives of the Vy Tribe (London, 1849), iii, iv.Google Scholar

³⁸ See Hair, , “Notes,” 36–49Google Scholar; and, for Koelle, Sigismund Wilhelm, Hair, P.E.H., “Introduction” to reprint of Koelle, S.W., Outlines of a Grammar of the Vei language, Together with a Vei-English Vocabulary, and an Account of the Discovery and Nature of the Vei Mode of Syllabic Writing (London, 1854).Google Scholar A preliminary version of the grammar had been published in 1853, but this was without the account of the discovery and the “Specimen of Vei writing” (see Hair, , “Notes,” 47Google Scholar).

³⁹ In 1856 Wilson referred to his discovery of the Vai script in 1834 and it being reported in the Missionary Herald, but he also reprinted a specimen of the script taken from Koelle, , Outlines, 251Google Scholar (Wilson, , Western Africa, 95Google Scholar).

⁴⁰ Two Monrovia publications, [Jangaba, S.Johnson, M.], Traditional History, Customary Laws, Mores, Folkways and Legends of the Vai Tribe (1954), 49–54Google Scholar, and a booklet on the Vai script published by a committee of Vai speakers, The Standard Vai Script (African Studies Program, University of Liberia, 1962 [note that the first page occurs in different copies in variant forms, without explanation])Google Scholar contain accounts of the invention (purportedly based on “oral traditions,” but largely following Koelle), which claim that Bukele and his associates composed the script at an earlier date, the 1954 publication giving 1819 and the 1962 one 1814 or 1816. Thus they echo the claim of some early accounts that Sequoyah began work on a script for the Cherokee in 1809, in each case probably an attempt to exclude missionary influence and elevate “independent invention.” Be that as it may, in 1836 a missionary reported that a school included “a Vey man, he writes the characters recently invented among his countrymen:” AR 12(September 1836), 280.

⁴¹ Bandakoro, which was “built” only in the mid 1840s (Koelle, , Narrative, 25Google Scholar), was later destroyed during Vai civil wars (perhaps more than once) and does not appear on the modern map (“Liberia,” 1:500000, U.S. Geodetic Survey, 1957). Jondu was earlier destroyed during an invasion of the “Guras,” that is, the Gola, the ethnicity to the northeast of the Vai. But an important town of that name exists today and appears on modern maps. Probably the earlier site was later resettled, but since Jondu means “slave house” (Koelle, , Outlines, 20Google Scholar), there could have been several towns of this name. However, Holsoe states categorically that the Jondu in question is the town now mapped in Gawula, a Vai territory to the east of Lake Piso, and he contends that the text written by Bukele is in the Gawula dialect of Vai. Holsoe, Svend E., “An Early Vai Manuscript from Liberia,” African Languages 2(1976), 54–55Google Scholar, map. Hoisoe further states that the town of Bandakoro is now abandoned and he maps the site fairly close to Jondu, but since the name means “under the cotton trees” there may have been other, perhaps earlier sites with that name. Koelle's account of his journey to Bandakoro, by canoe and overland (Koelle, , Narrative, 12, 14–18, 24, 28Google Scholar), is unhelpful in respect of precisely locating the Jondu and Bandakoro of that time. The earlier part of his course was parallel to the coast rather than directly into the interior, as the account fails to make clear, and his River Bisuma, said to be “eight miles wide” but “quite stagnant” and at points shallow, is the largely salt-water Lake Piso (admittedly a body of water which when Koelle traveled through it at the height of the dry season, in February-March, may have been full of sandbanks, making it seem more like a river). From Datia/Latia to “Da” his canoe then seems to have followed the lake edge round to the south side, before entering a minor stream. But no “Da” appears on the maps consulted, and Vai da/la means “entrance.” However, since two miles up this stream another destroyed town “Dshoni” was passed, and assuming that the present-day town of Johni stands on the same site, the stream is the “Johny/Jonni Creek” shown on two maps of 1890 and 1906 (in Büttikofer, J., Reisebilder aus Liberia [Leiden, 1890]Google Scholar; Johnson, H.H., Liberia [London, 1906]Google Scholar). From this stream a journey on foot of an estimated 4 to 5 miles took the party to Bandakoro. Yet it is odd that, while this route would have taken Koelle past the site of the destroyed or already resettled town of Jondu, he fails to note this. It has to be added that the two maps of later date show a “Bandacoro” somewhat further in the interior (east of present-day Gohn), and an alternative route there, up River Mafi and a side stream, Japanca Creek, with seemingly a much shorter distance to walk. But perhaps this route was closed to Koelle's party because of war. Koelle's total journey was one of some 15 to 20 miles, but Bandakoro must have been only some 10 to 15 miles from the sea by a direct route.

⁴² In historiographical terms it is extremely fortunate and valuable that an account of the invention was procured from the principal inventor himself, yet this evidence must be subject to certain reservations. The first is that Bukele spoke some fifteen years after the invention, a period during which he had had time to reflect on its significance, not least in the light of his locality's recent political upheavals and cultural developments. Perhaps even more importantly, we receive his story at second hand. Koelle was an excellent scholar, and although a missionary with an agenda for selecting queries and interpreting responses, his account bears little obvious evidence of commitment-reshaping of his interviews with Bukele. However, we are not fully informed about the mechanics. In 1849 Koelle spent only seven weeks in Vailand, for four weeks of which he was detained on the coast, and thus spent only three weeks with Bukele. “But I had not been thus engaged for more than a week, when I was seized by a violent attack of fever”—and as soon as he had recovered, Koelle left Bandakoro. The period of detention on the coast “I employed in making myself acquainted with the Vei language, which afterwards assisted me much. … For this purpose I had to hire a man from one of the neighboring towns who understood some English.” (Koelle, , Narrative, 8–9Google Scholar). On his second visit, in 1850–51, he first took down texts phonetically, but “I did not understand what I was writing” (Koelle, , Outlines, iiGoogle Scholar). It seems unlikely that when, after only four weeks' study of the language, he interviewed Bukele he had more than a very limited command of Vai; therefore problems of communication must have arisen. The text does, however, give the impression that the two could converse fluently. It is possible, although perhaps not very likely, that Bukele had sufficient command of broken English to understand Koelle's questions and make Koelle understand his responses, if not fully, at least to a reasonable extent. But it is more likely that they conversed in part through an interpreter. Whereas Koelle explained the amount of English of his 1850–51 informants (some had none), and recorded that translation from texts was done “with the aid of one who understands some English,” he stated nothing about an 1849 interpreter. It is true that he claimed to have left Sandbeach with only a wounded soldier and two boys in the canoe, but later he refers to a servant accompanying him (Koelle, , Narrative, 12–13, 16Google Scholar). Possibly the servant and one of the local Vai could converse in an intermediate language, and the servant then told Koelle, in English, what had been said. Either way—Bukele and Koelle attempting each other's language or through an interpreter—the possibility of misunderstanding over details cannot be ignored. While Koelle might well have checked Bukele's responses as far as he could, some small degree of error in the details of his account must be regarded as almost inevitable.

⁴³ The Appendix of 30 pages, written from “Fourah Bay” and dated June 1849, was inserted in only a few copies: we are indebted to the Library of the School of Oriental and African Studies, London, for providing us with a photocopy. The Appendix contains a free translation of three Vai script texts which Koelle brought back to Freetown. One of these texts is a short autobiographical piece composed by Bukele himself (Appendix, 2–6), but it does not refer to the script invention. Part of the text, in the script, phonetic transcription, and English word-by-word and free translation, appeared in Koelle, , Outlines, 241–52Google Scholar (although the script text was misleadingly described only as “A Specimen of the Vai Writing”); and again, in phonetic transcription and German word-by-word and free translation, with linguistic annotation, in Steinthal, H., Die Mande-Neger-Sprachen (Berlin, 1867), 274–79Google Scholar; and more recently, in original and modern script characters, phonetic transcription, and English word-by-word and free translation, in Holsoe, , “Early Vai Manuscript,” 36–52.Google Scholar

⁴⁴ Koelle did not see before publication the 1849 changes, which were mainly stylistic and probably the work of Henry Venn, the CMS Secretary. Koelle expressed satisfaction with the revised text, but offered, for a second edition, some changes of his own, mainly to the printed forms of Vai characters: CMS Archives, Section IV, Africa Missions, Part 5 (West Africa), Sierra Leone 1820–1880—examined in microfilm at the Schomburg Center for Research in Black Culture, New York Public Library (hereafter CMSA), CA1/0135/3. It might be noted that an earlier brief report, written by Koelle in May (CMSA, CA1/0135/2), was unchanged when published in the Church Missionary Intelligencer (09 1849), 103–04Google Scholar and map—the map unfortunately unhelpful in respect of Koelle's journey. Significant changes between the 1849 manuscript and booklet text are indicated in the cited passage in our text. Koelle was in England and Germany during 1853 and 1854, so was in a position to see through the press his grammar of Vai, in particular the enlarged 1854 version which contained the abbreviated account.

⁴⁵ For details of the 1889 changes, which are probably not significant, see note 59 below.

⁴⁶ Koelle, , Narrative, 21–24.Google Scholar Approximately one-half of the Narrative (“a short extract of the above-named pamphlet”) was reprinted as an appendix to Koelle's, Outlines, 229–40Google Scholar, the passage quoted above on 235–37. But many verbal changes from the 1849 version were introduced (perhaps by another hand, possibly that of Edwin Norris, the leading authority of the day on African languages, in order to improve instances of Koelle's slightly awkward English, e.g. “they early came” becomes “they came early,” 19). There were also a number of small textual changes, in the form of additions and deletions, which were presumably from Koelle's hand. Significant textual changes in the 1854 version are indicated in the cited passage in our text.

⁴⁷ Since Bukele had died in 1850, the 1854 version puts references to him in the past tense throughout the passage. We have not indicated the later instances.

⁴⁸ The 1854 addition was presumably by Koelle and came from rereading his notes. In 1826 Agent Ashmun wore a “thick coat,” to prevent ague, when the temperature fell “as low as” 74° (Gurley, , Ashmun, 126Google Scholar); and in 1833 it was reported that in Monrovia “the most respectable inhabitants” wore cloth coats, and that Governor Mechlin had stated that “when he travels he wears his coat and overcoat:” AR 9/3 (May 1833), 94. This distinction, between the unclothed heathen and the fully-covered civilized, is still a feature of belief and practice among the Christians of Sierra Leone and Liberia (as the distinction exists also, but in another form of clothing, among West African Muslims). The “long coat” mentioned by Bukele is a confirmatory indication that the dream figure was a non-Vai.

⁴⁹ We may speculate that the book in the dream was an image reflecting the presentation to the Vai of either the Koran or the Bible, or both. Muslims consider Christians (and Jews), like themselves, “peoples of the book,” as distinct from the heathen, who lack a written revelation. Later in Koelle's account, “the book” stands for a letter (epistle). The term “book” can also have a wider connotation. When in 1834 a missionary in Liberia sought permission from a chief to evangelize, the interpreter, a trader, told the chief “we wanted them to ‘sabby’ book all the same as ‘Merican people’” (HAM, 561). The phrase “to learn/teach book,” meaning to learn/teach reading and writing, is still used in West Africa today, when broken English is employed.

⁵⁰ For a discussion of the indigenous, Muslim, and Christian taboos in Bukele's account of the invention as reported by Koelle, see Dalby, , “Indigenous Scripts,” 12.Google Scholar

⁵¹ This, taken together with the later statement that the book came from God, resembles the Islamic belief that the Koran is a copy of an archetype, a book (of divine wisdom and will) existing in heaven, as revealed to the Prophet and by him dictated to followers. A version of this belief was expressed by the Vai ruler to whom the inventors revealed their script: “this was most likely the book, of which the Mandingos (who are Muhammadans) say, that it is with God in heaven, and will one day be sent down upon earth” (Koelle, , Narrative, 24Google Scholar). In 1833/34 Wilson and Wynkoop met a Susu Muslim who “had a book at home which he regarded as sacred … given [him] by two old Mandingo men, let down from heaven in a brass pot:” MH 30/9(September 1834), 335. Koelle averred that “Muhammadism had no hand in the invention of the Vei mode of writing” but it was used “afterwards [Koelle's emphasis] for Muhammadan purposes” (CMSA, CA1/1035/7). Again, in print he claimed that, although “Muhammadan influence on the contents of the accompanying manuscripts is undeniable, and that much Muhammadan nonsense is contained in them; nay, they may be translations from professed Muhammadan Arabic books, but this has nothing to do with the invention of the mode of writing itself” (Koelle, Narrative, Appendix, 1). Nevertheless, it is arguable that Islamic influences played almost as large a role in the invention and dissemination of the Vai script as Christian missionary influences, but this is not the subject of the present paper.

⁵² The missionary Pinney noted that, when visiting villages in Liberia in 1833 and 1834, in order to satisfy the young men seeking education and demanding that it commence immediately, he “began on the spot to draw letters in the sand:” AR, 12/8(August 1836), 248.

⁵³ Note that, whereas the earlier encouragement “to write any Vei words in the same way, in which the book was written” might denote an alphabetic script (as in the Koran or Bible), this reference to a single “sign” for na predicates a syllabic script.

⁵⁴ The source of the additional clause, “when quite a little boy,” is unclear. It was not among the changes Koelle suggested for a second edition of the 1849 booklet.

⁵⁵ Later commentators have tended to assume that the slave traders on the Vai coast were Portuguese, for instance, the 1962 Monrovia account has Bukele going “to Ghana [sic]as a steward to a Portuguese” (The Standard Vai Script, unnumbered first page). But in fact most of the slavers were non-Portuguese, such as the Spaniard Pedro Blanco and the Frenchman Theophilus Canot/Conneau; see a list of Gallinas slavers in Jones, Adam, From Slaves to Palm Kernels: A History of the Galinhas Country (West Africa) 1730–1890 (Wiesbaden, 1983), 43.Google Scholar Since Bukele's employment must have occurred well after 1807, it is unlikely that he was employed by British slavers, or that on such business Bukele visited Freetown or the downcoast ports controlled by the Liberians. But he may have visited these places when his employers were engaged on “legitimate” trade. Moreover, the slavers included North Americans, from whom Bukele might have acquired some English. However, his alleged employment as a servant to non-Africans might be misleading, since his autobiography shows him operating in a trading capacity within African networks. At the uncertain date of his father's death he had been sent by his father from Jondu to Sowii (Sowili/Sewulu) on River Mani (a little to the north of Lake Piso) “to sell camwood,” and he subsequently proceeded to River Gallinas where his sister (or step-sister) was a wife of the king's son, with whom, in a death-announcement ritual, trade goods were exchanged—a slave, cloths, rum, a gun and gunpowder (Koelle, Narrative, Appendix, 44–45; Holsoe, , “Early Vai Manuscript,” 51Google Scholar). While Bukele might have served an apprenticeship to a non-African trader, he patently was a member of a respected and influential trading family. In the 1880s his son was chief of Mendo/Mando, a town in the southeastern corner of Lake Piso (Büttikofer, , Reisebilder, 2: 243Google Scholar).

⁵⁶ Since, according to the same account, Bukele had previously become acquainted with the written word, his question is merely rhetorical.

⁵⁷ In the 1828 Missionary Herald account of the invention of the Cherokee script, Sequoyah's young comrades similarly puzzle over how whites communicate at a distance—an intriguing coincidence. In both cases this late puzzlement (after a long period of contact and encounter) seems implausible and smacks of white interpretation. It is conceivable that Koelle's account of the invention of the Vai script in this aspect was influenced by his having read, perhaps in an intermediate source, the missionary account of the Cherokee script invention, but it does not seem very likely.

⁵⁸ Cf. “Poro country (very likely originally equal to Portuguese country, but now to ‘white man's country’ in general)” (Koelle, , Narrative, 19Google Scholar). In his Vai vocabulary Koelle gave a fuller and probably more significant explanation of the term “Poro.” “Probably a corruption of Portuguese … it became a denomination for white men in general. It is now applied to Europeans and Americans, and by way of politeness also to those Negroes who have some education and are more civilized than the natives of the country … they have often told me, we know very well that they are not real Poros, but we call them so because they have been in white man's country and like to be called so” (Koelle, , Outlines, 207Google Scholar). Probably the term derives from pong “distant, far away” + moenu “people,” i.e., “people from far away.”

⁵⁹ In Koelle, S.W., Mohammed and Mohammedanism (London, 1889), 63Google Scholar, Koelle (who had latterly lived for many years in Turkey), when discussing the dream revelation of Muhammad, compared it with the dream revelation of Bukele, which he recounted at length. Although he mentions the 1854 text, at points this very late version modifies and even contradicts the earlier ones. It is difficult to know what to make of these changes, since they may have arisen through mere forgetfulness some forty years after the original episode. For what the changes are worth, in 1889 Koelle stated that (a) Bukele encountered a missionary who soon left the country “when a little boy,” thus supporting the 1854 version; (b) Bukele learned a few Scripture passages in English but “not how to read and write,” this negative statement being compatible with the earlier statement that “he learnt to read for about three months” only if we suppose that both meant that Bukele began lessons in reading and writing but did not progress very far; (c) Bukele, when a man, had a dream in which there appeared “the white teacher of his childhood,” a claim about the dream figure lacking in the earlier versions; (d) Bukele, when he awoke, not only remembered how the dream figure wrote signs in the sand but “distinctly remembered many of the signs taught him” and so was able to form his syllabary, strengthening what is only implied in the manuscript version by the inclusion of two actual Vai characters.

⁶⁰ “Rora,” otherwise Ndlε (“the correct phonetic spelling,” Stewart, , “Early Vai Script,” 3Google Scholar), was the “book name” of Kali Bara.

⁶¹ The script text was published in [Norris, E.et al. (?)], eds., The Book of Rora (London, 1851)Google Scholar—see Stewart, Gail and Hair, P.E.H., “A Bibliography of the Vai Language and Script,” Journal of West African Languages 6(1969), 114Google Scholar — and an English translation had already appeared in Koelle, Narrative, Appendix [see note 43 above], 8–14), there described as “the Manuscript of Kali Bara, who, when writing, calls himself Rora.” Based on the above works, the text, in phonetic transcription and German word-by-word and free translation, again appeared in Steinthal, , Mande-Neger-Sprachen, 280–96.Google Scholar After the script manuscript resurfaced in the 1960s, and in criticism of the allegedly misshapen forms of certain characters produced by earlier non-Vai scholars in printed works, the part of the text relating to the script invention, in revised phonetic transcription and English word-by-word and free translation, appeared in an unpublished paper, Stewart, , “Early Vai Script,” 28–29.Google Scholar But for the earlier part of the text we are still dependent on Koelle's 1849 translation (made at “Sandbeach,” Cape Mount: Koelle, , Narrative, 33Google Scholar). It might be added that the third script text, entitled by Koelle “the Manuscript, once belonging to king Goturu, in the Vei country,” contains largely moral apothegms of a strongly Islamic flavor, apparently proceeding from Goturu himself, with odd sections describing the king's wars (Koelle, Narrative, Appendix, 14–30).

⁶² Yet in 1850–51 Koelle reported that “Kári Bára Ndoré Wanó, the author of one of the manuscripts” spoke “only a few words of English” (Koelle, , Outlines, iiGoogle Scholar). For the name subsequently given as “John,” see ibid., 20 “John (pronounced Dsáni).”

⁶³ Stewart, , “Early Vai Script,” 29–30Google Scholar (the names, given in the special characters of the International African Alphabet, are cited here in standard characters). Stewart transcribed the passage from the manuscript, and the English translation is hers; both therefore differ from the ones offered by Koelle and used by scholars previously. We have, however, changed the translation of “polo-monu” from “Europeans” to “Those-From-Afar,” in view of the discussion of the term that follows in the text. Koelle's slightly different translation reads as follows. “God taught us, to know this book. All of you know this. Yet there exists some difference respecting the writing of it. Rora and Dshara Lau and Dshara Kali and Salu Tubaku and Fa Gbase and Doalu Gburo [this is Bukele], these six men lived with John, to learn English in Dshondu. At that time began my Father Worogbe to like books. And the people said: The Poros have long heads; nobody has such a long head, as the Poros. But some of our people did not believe this. Then said Doalu: why do you say, what I maintain is a lie? Can any Vei-man write a letter and send it to his friend and could he read it? — Then it was day-light and all the young men went to Mr Gang's [?] big kitchen and began this book palaver. They wrote letters and sent them to each other. It is done” (CMSA, CA1/1035/16).

⁶⁴ Koelle, , Narrative, 22.Google Scholar

⁶⁵ Kali Bara was Koelle's “landlord” during his stay at Bandakoro in 1849 (that is, the owner of the house in which the visitor lived). The earlier part of Kali Bara's manuscript account was copied by the author himself (on to paper Koelle provided in order to take away the copy) but the final passage, dealing with the invention, was copied by Koelle (Koelle, , Narrative, 18–19Google Scholar). Stewart speculated that this passage was not part of the original book, but was composed by Kali Bara for the occasion, and that this might explain “the elliptical nature of the story and the failure to mention Duwalu's dream—perhaps it was haste or perhaps reticence” (Stewart, , “Early Vai Script,” 18–19Google Scholar). It may be that Kali Bara thought it right to leave the telling of the story of the dream to Bukele, his cousin and fellow townsman. But Koelle disbelieved Kali Bara's claim that he too had had an inspirational dream, suggesting that he had doubts about Kali Bara's veracity and, while he wrote at length about Bukele's good character, his only comment on Kali Bara was that his head wife “does not yet appear to be very happy in his company” (Koelle, , Narrative, 29Google Scholar). The earlier part of Bukele's manuscript relates only to the death of his father and the consequent proceedings, at an unstated date which Holsoe, from references to two known kings, places between 1835 and 1843 (Holsoe, , “Early Vai Manuscript,” 54Google Scholar). However, if the “Bandakolo” which is mentioned was the one visited by Koelle, as Holsoe states, and if Koelle was correct in stating that the town had only been founded “about five years” before his visit in 1849 (Koelle, , Narrative, 25Google Scholar), then the manuscript dates from the mid-1840s.

⁶⁶ A Vai scholar, then an official of the Cultural Affairs section of the Liberian government, S. Jangaba M. Johnson, has suggested that Kali Bara's account relates in time, not to Bukele's invention, but to a later episode, its revival after a check to the dissemination of the original script, caused by a war (Stewart, , “Early Vai Script,” 16Google Scholar, citing a personal communication from Johnson and his 1954 Traditional History, 54). We believe that the account does relate to the invention, but are inclined to suspect that it was written after Jondu was destroyed, and that it might reflect political circumstances in which it was appropriate to enhance the role of the young men and perhaps to play down the role of Bukele, an older man whom Koelle portrays as a somewhat idealist thinker, not—or no longer—a practical man of action.

⁶⁷ Duwalu Bukele is termed Nfa Duwalu Worogbe, “Nfa” being a Vai title given to a more senior male and Worogbe a patronymic. It is worth noting that in Bukele's account the name of Kali Bara comes third in the list of co-inventors, but in Kali Bara's account he names himself as the first in the list of inventors.

⁶⁸ Dalby, , “Indigenous Scripts,” 9–10.Google Scholar The text does not actually say that those involved were “living at” Jondu, only that they were there learning English (Koelle's translation having been modified by Stewart).

⁶⁹ Daniels, /Bright, , Writing Systems, 579.Google Scholar The spiritual aspect of the invention and use of the script was confirmed by Bukele himself, when in 1849 he stated that “once I thought to find God in our book-palaver” (Koelle, , Narrative, 26Google Scholar).

⁷⁰ Koelle, , Narrative, 22.Google Scholar

⁷¹ Dalby, , “Indigenous Scripts,” 10.Google Scholar

⁷² American Anti-Slavery Society, Examination of Mr. Thomas C. Brown, a free colored citizen of S. Carolina, as to the actual state of things in Liberia in the years 1833 and 1834, at the Chatham Street Chapel, May 9th & 10th, 1834 (New York, 1834), 29.Google Scholar For a similar explanation, see Gurley, , Ashmun, 134.Google Scholar See note 58 above for a similar use of the term “Poro” to denote the Afro-American settlers as well as whites.

⁷³ Holsoe, , “Stimulus Diffusion,” 23.Google Scholar Strictly, po moenu is plural, “civilized people,” not “civilized man.”

⁷⁴ Koelle, , Outlines, 237.Google Scholar

⁷⁵ Hair, , Notes,” 41–42.Google Scholar

⁷⁶ Jones, Adam, “Who were the Vai?,” JAH 22(1981), 159–78.CrossRefGoogle Scholar

⁷⁷ For what is known of the earlier history of the Vai in the Gallinas/Galinhas district see Jones, Front Slaves to Palm Kernels, chapters 2–6.

⁷⁸ American Baptist Foreign Mission Society, Africa Correspondence 1814–1900 (hereafter ABapFMS) (seen on reel FM-81 of the microfilmed archives of the Baptist Historical Society, Valley Forge, Pennsylvania), correspondence of Lott Carey 1826–1828, letter from Carey and C. Holton to Baptist Board of 16.2.1826.

⁷⁹ AR, 3/8(October 1827), 245–46; cf. Gurley, , Ashmun, 329Google Scholar; MH 23/12(December 1827), 391. But contrary to the view that Islam was spreading widely, in 1834 Wilson and Wynkoop opined that the Vai at Cape Mount “have no religion, except a few who have embraced the Mohammedan faith:” MH 30/9(September 1834), 336. Governor Ashmun, who visited Cape Mount in March 1827, reported of the Islamic teachers that “they confine their principal efforts to the instruction of their youth and children. These they teach to write the Arabic in a fair character, and to read and understand the Koran and other books, of which they always possess several in manuscript.” The adult males of the Vai may well have resented this literacy of local children in an exclusive script, which may have helped to encourage them to produce a script of their own. Islam was conveyed from the interior, often by visiting Manding traders (Jones, From Slaves to Palm Kernels, chapter 6), who were able to converse with the Vai with fair ease, Manding and Vai being closely related languages. Governor Russworm of Liberia reported in 1829—“Yesterday I was visited by two Mandingoes who wrote Arabic with great care … I showed them a Greek book [the Greek New Testament?], which they desired me to read, and they would read Arabic:” MH 26/6(June 1830), 188. The anecdote testifies to the contemporary missionary interest in, for them, exotic languages. Islam continued to advance and in 1849 Bukele, the principal inventor of the script, told Koelle that, though not spiritually satisfied, he had “prayed after the Mandingo [i.e., Muslim] fashion these seven years” (Koelle, , Narrative, 26Google Scholar). Today the vast majority of the Vai acknowledge Islam.

⁸⁰ MH 30/6(June 1834), 215; AR 10/6(August 1834), 178. It was probably the Islamic presence at Cape Mount that encouraged Wilson and Wynkoop to recommend that a mission of their organization be established, not at Cape Mount, but far to the east, at Cape Palmas, among exclusive “heathen.”

⁸¹ AR 1/5(July 1825), 154–55; American Baptist Magazine (hereafter ABM) 5(1826), 241.Google Scholar John spoke broken English, as recorded in a long pious speech he made. He “received his [religious] impressions about three years ago at Sierra Leone, and while there he got the knowledge of his letters. After about three months' advantage of schooling, his relations called him from Sierra Leone to Cape Mount. … He however took along with him a spelling book, and he continued … trying to spell [on his own, without a teacher]” (HAM, 441). In June 1827 John was said to read the New Testament “middling intelligently,” but it was claimed, wrongly, that he had learned to do so only in the recent period of six weeks. A letter from him in perfect English and spelling was clearly written by someone else at Cape Mount, probably an Afro-American trader: ABM 7 (1827), 304. While in Freetown, John had come under the influence of Hector Peters, an elderly Baptist and a “Nova Scotian,” that is, an Afro-American settler of the 1790s, a freeman originally from Charleston: Fisher, Miles Mark, “Lott Cary, the Colonizing Missionary,” Journal of Negro History 7(1922), 395CrossRefGoogle Scholar; Wilson, Ellen Gibson, The Loyal Blacks (New York, 1976), 85, 339.Google Scholar The Freetown community in the mid-1820s included not only a large number of “Nova Scotians,” but also those Afro-American colonists who had settled there after the failed expeditions of 1820 and 1821 and had subsequently declined to join the 1822 settlement in Liberia. Presumably it was these contacts in Freetown which encouraged John, after he was recalled to Vai country, to visit the Afro-Americans at Monrovia for baptism. When he afterwards evangelized at Cape Mount, his first convert had the English name George Peter (Peters?), perhaps taken on conversion and the surname conceivably in tribute to John's Freetown mentor: ABM 8(1828), 54.

⁸² Discussing all the African peoples with whom the Liberians were in contact, Ashmun was categorical. “They have not in general use, nor am I certainly informed that they use at all, any paintings or hyeroglyphical representations whatever, for recording or communicating facts” (“Traits of the African Character,” dd 20 11 1824: AR 1/2[April 1825], 59).Google Scholar He was perhaps thinking of the pictograms allegedly used among certain North American peoples. He went on to note the writing within the Islamic amulets sold locally, which he termed “the Mandingo writing, which still preserves the distant and defined likeness of the Arabic character,” although neither the writers nor those who bought the amulets pretended to understand the language (ibid., 61). For a lengthy discussion of whether the West African indigenous scripts, including the Vai script, owe anything to a variety of “sub-Arabic” scripts, some found for instance occasionally in amulets; see Dalby, “Inspiration and Design.”

⁸³ AR, 3/9(November 1827), 259; also in Gurley, Ashmun, Appendix, 81 (Ashmun further spoke of the Vai as “active, warlike, proud and … deceitful. The slave traffic has furnished them with their principal employment, and proved the chief source of their wealth, to the present year, when it is believed to have been broken up entirely and forever. … Three-fourths of the population are domestic slaves, now engaged in a civil war with their masters for an extension of their privileges”). When Cary preached at Monrovia in 1827 among those attending were a number of “natives from Cape Mount … who understand English:” ABM 8(1828), 54.

⁸⁴ ABM 7(1827), 305 (letter of June 1827).

⁸⁵ Report of the American Society for Colonizing the Free People of Colour of the United States (Washington, 1833), 3.Google Scholar The Cape Mount school was also mentioned in relation to early biographies of Lott Cary that appeared in Freeman, Frederick, A Plea for Africa, being Familiar Conversations on the Subject of Slavery and Colonization (Philadelphia, 1837), 317Google Scholar; Gurley, Ashmun, Appendix, 159.

⁸⁶ It is fair to observe that not only the Vai were interested in literacy. In 1834 it was noted that “along the whole coast [between Cape Mount and Cape Palmas] … we uniformly found the people desirous of schools. … We had multiplied, pressing solicitations to receive their sons at Cape Palmas [where a school was being established] and educate them there:” AR 10/6(August 1834), 181.

⁸⁷ The Vai had some reason to be wary about the Liberians. After a defensive war at Monrovia when the early settlers killed many local Africans, both Ashmun and Cary led a number of successful military assaults on slave ports up and down the coast — justified on the abolitionist grounds that the slave trade had desolated the region — which killed more Africans, allies of the slavers.

⁸⁸ It is not implausible that the Sherbro settlement had a school which attracted children from a distance, as the Monrovia school was to do later, especially if the children were those of prominent men with whom traders like John Kizell (see note 114 below) had contact.

⁸⁹ Given their knowledge of two different scripts, the Roman and the Arabic, it is most likely that the Vai did not regard their invention of their own different script as in any way out of the ordinary. The adoption of a syllabic form, which has so impressed later scholars, was contingent, and not intended to score any cultural or political point, as occasionally suggested in latter-day writings, inasmuch as presumably the inventors were unaware of the rarity of the form in the modern world. If any of the inventors claimed the better functionality of a syllabary for the Vai language, a case considered by later linguists, there is no record of any such discussion, and it is inherently unlikely.

⁹⁰ The idea of a syllabary certainly diffused across West Africa later. The syllabic scripts of the Mende, Loma, and Kpelle, all devised in the vicinity of Vai country in the first half of the twentieth century, were stimulated in design by the syllabic organization of the earlier Vai script. The Bassa alphabet, introduced in Liberia in this same period, was constructed by a western-educated Bassa who was familiar with the Vai script. Further down the coast, there is evidence suggesting that the Bamum syllabary (and thus the later Bagam syllabary inspired by it) was influenced by the syllabic blueprint of the Vai script. A Bamana syllabary devised in the 1930s has been recorded in Mali, and this script too may have been influenced in its creation by the continuing success of the Vai script and the presence of Vai traders. A noteworthy point is that in all instances, while the African inventors were stimulated by knowledge of earlier syllabaries, they chose to fill out their own scripts with novel characters.

⁹¹ Nichols, John D., “The Cree Syllabary” in Daniels, and Bright, , Writing Systems, 599.Google Scholar

⁹² For the influence of the Cree syllabary on other scripts see Young, Edgerton R., By Canoe and Dog Train among the Cree and Salteaux Indians (New York, 1891), 57, 142Google Scholar; Nichols, , “Cree Syllabary,” 599–611.Google Scholar

⁹³ It is not improbable that printed reports of the success of the Cherokee script were actually available and read in Liberia. Information on the Cherokee script was covered widely in the American, and some of the European, newspaper press, as well as in Anglophone missionary journals. In 1826, at Monrovia, “a room had been set apart in the wing of the old Agency house, for the Colonial Library, consisting of one thousand two hundred volumes, covered accurately, labelled, and systematically arranged in glazed cases with appropriate hangings. Files of American newspapers were here also preserved, and it was intended to render this apartment, both a reading room and a museum, for African curiosities” (Gurley, , Ashmun, 330Google Scholar).

⁹⁴ The first Baptist missionary, the Afro-American Lott Cary, arrived in Liberia in 1822 and died in 1828; his immediate successors were Afro-Americans; Crocker and a colleague were the first white Baptist missionaries.

⁹⁵ Medbery, Rebecca B., Memoir of William G. Crocker, Late Missionary in West Africa Among the Bassas, Including a History of the Bassa Mission (Boston, 1848), 112.Google Scholar

⁹⁶ “2 months ago” puts the inception of the idea of a syllabary in December 1835. Crocker and his mission partner Mylne were at Millsburg, 22 miles from Monrovia, for most of that month, “their employment… chiefly the study of the Bassa language, which is more generally spoken in that vicinity than any other:” ABM 16(1836), 132.

⁹⁷ ABapFMS, 80/28, correspondence of W.G. Crocker 1837–1838, journal entry of 8.2.1836. Cf. “Last week, while looking at the Cherokee alphabet, it occurred to me that an alphabet on similar principles might be constructed for the Bassa language. I immediately set about making one, and have been highly prospered by God; so that we [have] a syllabic alphabet by which I believe we can spell all the words in the language. I believe the same principles may be applied to other native languages, and hope that much good may come to Africa from it” (Medbery, , Memoir, 111Google Scholar, dated to 16.2.1836). (Comparison with the manuscript text of Medbery's journal shows that—as was not uncommon in nineteenth-century biographies—the printed journal entries often summarize, conflate, and misdate the original material. In the preceding quotation the reference to “last week” is wrongly attached to the date 16.2.1836, which misleads.) See also “Extracts of a letter from Dr. E. Skinner, Colonial Agent, to the Rev. R.R. Gurley, Secretary of the American Colonization Society, dated Monrovia, April 1836,” AR 12/6(June 1836), 174. In 1826 Ashmun had suggested creating for the Bassa “a new alphabet” with one character for each sound — this would therefore not have been a syllabary (Gurley, Ashmun, Appendix, 84, 88). In his extensive reports Ashmun does not appear ever to have mentioned the Cherokee syllabary or the concept of a syllabary.

⁹⁸ Medbery, , Memoir, 138.Google Scholar

⁹⁹ The arrangement of the Bassa syllabary shows some resemblance to Worcester's arrangement of the Cherokee syllabary, but there is no graphic resemblance between the characters of the two, nor between the characters of the Bassa syllabary and the Vai syllabary. For the single extant copy of the Bassa syllabary, see ABapFMS/Crocker.

¹⁰⁰ Peck, Solomon, “History of the Missions of the Baptist General Convention,” in HAM, 563.Google Scholar Crocker later reported that one of the “native boys” had been able to “read and write in the syllabick alphabet” (ABapFMS/Crocker, letter to Baptist Board of 12.4.1837).

¹⁰¹ To the best of our knowledge, no vocabulary of Vai was in 1836 available in print (see Stewart, and Hair, , “Bibliography,” 109–24Google Scholar). However, in 1834 Baptist and Methodist missionaries had worked, at least for short periods, at Cape Mount, and might well have collected a Vai vocabulary—as indeed may the administrators and missionaries who visited there earlier. A possible transmitter of a Vai vocabulary was A.W. Anderson, for whom see note 104. It might be argued, that although Crocker refers only to the Vai language and not to the Vai syllabary, his introduction of the former immediately after referring to the Cherokee and Bassa syllabaries indicates that at the back of his mind was a connection, that is, via the Vai syllabary. But such conjectural mind-reading is hardly firm evidence.

¹⁰² Yet Crocker, on arrival in Liberia in 1835, met briefly, and took over the house of, Pinney, the missionary who in 1834 had informed Wilson and Wynkoop about the existence of the Vai script (HAM, 560). We have to admit that Crocker's apparent ignorance is not easy to explain.

¹⁰³ Perhaps significant only in the realm of coincidence, a dubious connection between the Vai and Bassa syllabaries can be traced, which is at least of interest in that it challenges Crocker's judgment that the Bassa were “unambitious” and had “gone on for ages in the same track.” As noted by Dalby, (“Indigenous Scripts,” 22Google Scholar), the “Book of Rora” states that Kali Bara (Rora/Ndole) was the child of a Vai mother from Jondu and a Bassa father, and was born at Juring on River Gallinas. His father returned to Bassa country, and his mother to Jondu, where Kali Bara took a wife, and “whilst he lived there the Guturu war began.” This last statement may provide a date. In 1830, at the death of a paramount ruler, war broke out in Gawula between the heir and a challenger, Gotolo or Fantolo (Holsoe, , “Cassava-Leaf People,” 127Google Scholar). Perhaps because of the war, Kali Bara and his wife “went to the Gallinas for three years,” that is, perhaps up to 1833. Then, desiring a farm for himself, and having met a man from Bassa country, he decided to seek out his father. “On his journey he slept at Little Cape Mount (River Lofa) ten times” and thereafter “walked a long time”—conceivably via Monrovia and the local Bassa community—before meeting his father in Bassa country. Several place names are mentioned but have not been traced on the modern map (although one, Belewoi, has the shape of a Bassa placename, cf. Belefuenai, Feyawoie), and these localities were probably in the Bassa interior, “up country.” While there, Kali Bara met a whiteman, “James,” whom we have been unable to identify, perhaps a trader rather than a missionary. Apparently soon after his arrival, his father died and Kali Bara attended the appropriate ceremonies before returning to Jondu and to Gotolo's capital, Tienamani (Koelle Narrative, Appendix, 6–9). Within this account, the “Book of Rora” does not refer to the invention of the script, and although it seems to place Kali Bara in Jondu where the script was invented immediately before 1830 and again in the mid-1830s, it does not specify that he passed through Jondu on his journey from the Gallinas to Bassa country in, perhaps, 1833. The account of the script invention which appears at the end of the manuscript (as presented by Koelle) is not dated and does not attempt to link up with the autobiographical section. What can we conclude? Certainly that the Vai and the Bassa were not without contacts, both earlier and in the 1830s. Probably that Kali Bara knew of the script when he visited Bassa country. But that he proselytized for a Bassa syllabary and that this influenced Crocker, either when the latter was in the Monrovia district or at his mission station on the coast, seems very unlikely and is totally unevidenced.

¹⁰⁴ We discuss below the missionary possibly associated with the invention of the script, John Revey. The other missionary was A.W. Anderson, who described himself in 1835 as a Black, licensed to preach at the First Colored Church of Savannah. For “three or four months” in early 1834, on behalf of the Liberia Missionary Society, he was the teacher of a school “at or near Big Town” near Cape Mount, with “some success amongst the Vey tribe,” but was then withdrawn by the Monrovia mission authorities, apparently in part because of being accused of trading “in the interior,” the accusation coming from a resident trader (ABapFMS, correspondence of A.W. Anderson 1834–1844, letter to Anderson of 4.4.1834, letters to Baptist Board from Anderson of 11.5.1834, 14.1.1835). But he was later employed by the Baptist mission and in September 1835 he opened a school at Caldwell near Monrovia: ABM 16(1836), 92, 133. It appears that he was subsequently in contact with Crocker, since the latter noted in June 1836 that a specimen of the “Syllabic Alphabet … we sent home by Bro. Anderson” (ABapFMS/Crocker, letter to Baptist Board of 21.6.1836). The contact may have been made when Crocker was in Monrovia or Millsburg in later 1835, and Anderson's apparent interest in Crocker's Bassa syllabary may indicate that it was from this former Cape Mount missionary that Crocker learned about syllabaries and was encouraged to start work on a Bassa syllabary. However, if Anderson told Crocker about the Cape Mount experience, the letter's failure to mention — and his apparent ignorance about — the Vai syllabary is even more difficult to explain.

¹⁰⁵ “… the Veys … , I am told, have a method of writing, invented by themselves, by means of which they communicate with each other” (ABapFMS, correspondence of W. Mylne 1835–1839, letter to Baptist Board of July 1838).

¹⁰⁶ Baptist Missionary Magazine (new name of American Baptist Magazine, henceforth BBM) 17/6(06 1837), 133Google Scholar; Medbery, , Memoir, 145Google Scholar; see also Peck, , “Missions,” 563.Google Scholar Crocker abandoned the syllabary “on more mature consideration” and his new “orthography conformed to the principles generally observed by missionaries in applying the Roman alphabet to heathen languages.” Behind this abandonment of the syllabary was the view which had won the day in the United States, a view strongly pressed by the linguist John Pickering, that in all mission fields, including the Cherokee, the writing of newly-literate languages should be consistently in the Roman alphabet “in English letters:” BMM 17/5(May 1837), 119—a practical argument being that this cheapened printing. Crocker's mission partner Mylne noted that “we have read Pickering's plan of writing the Indian languages with deep interest. We have commenced the Bassa with it. We think that perhaps all things considered it may be found more suitable than the Syllabic Alphabet” (ABapFMS/Mylne, letter to Baptist Board of 12.09.1836; cf. ABapFMS/Crocker, journal 2.12.1836). In 1837 Crocker commended an eleven-year old boy who could “now probably write about anything he wishes on the principles of Pickering's orthography,” having been taught by a Liberian who had “made himself acquainted with Pickering's principles” (ABapFMS/Crocker, letter of 12.04.1837).

¹⁰⁷ BBM 17/6(June 1837), 133; Medbery, , Memoir, 220, 236Google Scholar; see also AR 13/3(March 1837), 83; (June 1837), 199; Peck, , “Missions,” 563.Google Scholar Crocker's Bassa spelling book was probably printed at Monrovia in 1837/38, but no copy has yet been traced. A second printing at Cape Palmas was in 1840 or 1841. His translation of two Gospels, completed in 1841, was printed at Edina in 1844: BBM 17/5(May 1837), 119; Bleek, W.H.I., The Library of Sir George Grey: Philology (2 vols.: London, 1858), 1/2:227–28.Google Scholar

¹⁰⁸ Report of Commander William Francis Lynch, In Relation to His Mission to the Coast of Africa, 1852–1853 (Washington, 1853), 37.Google Scholar A legend latterly current among the Bassa to the effect that, before the arrival of American settlers, a simple form of pictographic writing called “Vah” had been invented by a Bassa man (Dalby, , “Indigenous Scripts,” 33–34Google Scholar), might have its roots in distant echoes of Crocker's Bassa syllabary. In 1907 a Bassa alphabet was introduced in Liberia by a Western-educated Bassa on returning from the United States, and has continued in limited use. Its originator appears to have been inspired by the Vai syllabary rather than from any knowledge of the earlier Bassa syllabary.

¹⁰⁹ Peck, , “Missions,” 441Google Scholar; see also Medbery, , Memoir, 74.Google Scholar

¹¹⁰ Gurley, , Ashmun, 356–57.Google Scholar

¹¹¹ ABapFMS, correspondence of C. Holton, 1826, letter to Baptist Board of —, .01.1827. The letters also stated that “if clothing is purchased for Cape Mount Boys, it must be good or they will not wear it. They are the proudest nation on the coast. They had rather go naked, than not wear the best for this country.”

¹¹² In February 1826, when a Liberian who had traveled to Cape Mount “to restore his health” returned to Monrovia, he reported that “King Peter had prepared a house for a school,” and Cary then stated that, in response to “frequent and urgent requests of King Peter for commencing a school at Grand Cape Mount,” assistance had been promised and he himself expected to visit the locality shortly (ABapFMS/Cary, letter of Cary and Holton to Baptist Board of 16.02.1826). For various reasons, Cary was unable to go to Cape Mount or send a teacher to establish the school, before 1827 (ibid, letter of the same to same, 22.04.1826; letter of Cary to same, undated but 1828).

¹¹³ AR 4/2(April 1828), 40; Fisher, , “Lott Cary,” 406.Google Scholar “Big Town” was said to be “some thirty-five miles in the interior from Cape Mount,” but this was almost certainly a misdirection and an exaggerated distance inland. It is highly unlikely that a pioneering mission school with a Liberian teacher would at this date have been settled so far into the interior of virtually unknown territory. All immigrant settlements and mission stations in Liberia were for long confined to the coastal zone. Holsoe has suggested that “Big Town” represented the interior capital, Gohn Zolu Duma, of the local Vai ruler, Zolu Duma alias King Peter, Revey having reported a burial ceremony for the ruler from there (personal communication to K.T., 1.10.2000). However, Holsoe has noted that several “big towns” existed (Holsoe, , “Cassava-Leaf People,” 131Google Scholar: “one of Fanl's big towns”). The term “Big Town” was not recognized as a present-day name by a Vai informant, and could have been a descriptive term applied to several localities in the past, perhaps merely for the benefit of non-Vai. According to letters from Revey what took place at “Big Town” in February 1828 was “the burial of old king Peter's bones” and a funeral wake lasting many days (ABapFMS/Cary, letters of Revey to Cary 13.02.1828 and 28.02.1828). The present town with the name Gohn Zolu Duma lies 35 miles inland on a direct route to the coast but some 45 miles from Cape Mount following Koelle's route via Lake Piso. Other references described Big Town as “at Cape Mount,” or perhaps more specifically (since “Cape Mount” could indicate a district “near Cape Mount: ABM 15(1835), 114, repeated, 234. While in correspondence, Cary spoke of the “establishment at Cape Mount” and the need for a boat to travel up and down the coast, presumably between Monrovia and Cape Mount, and he complained that goods for the school had had to be transported all the way overland, which was expensive (ABapFMS/Cary, undated letter to Baptist Board of 1828). Cary supposed that Big Town contained “a population of at least 600:” ABM 7(1827), 304. It might be significant that when a boat was wrecked, a man escaped ashore “from the waves” and was then looked after by the Vai evangelist, John (Peck, , “Missions,” 441Google Scholar), for if John was at this time living at Big Town, this suggests that the locality was close to the sea. Although, curiously, the Liberian missionaries fail to specify exactly where Big Town lay—or otherwise the particular “Big Town” where the school existed—the issue appears to be partly resolved by a reference in a letter of 20 December 1828 from the Big Town of the school by Johan Handt, a German missionary discussed below, which placed it at 12 English miles from Cape Mount (Basel Mission Archives. Afrika Mission nach Liberia 1826–1832, 1829 Nr.3). This probably indicates the town of Mondo/Mando, at the southeastern corner of Lake Piso, which lay only 3 to 4 miles from the sea and could be approached by the lake. It had earlier been the capital of a coastal ruler, “King Grey” (Holsoe, , “Cassava-Leaf People,” 123Google Scholar). Jondu, where Bukele invented the script, and a possible site for the school (as we proceed to consider), was also some 12 miles from Cape Mount but only in a direct line crossing Lake Piso, overland it being a further 5 to 6 miles further on than Mondo; and it was not close to the sea, but some eight miles from it. Although Koelle traveled some 12 to 15 miles from “Sandbeach” near Cape Mount (probably later Robertsport) before passing near the site of Jondu, and while it may have been normal to approach the Gawula district by the route through Lake Piso rather than directly from the sea (with its surf beaches), nevertheless the Monrovia estimate of “35 miles into the interior from Cape Mount” was seemingly a gross exaggeration. A later writer claimed that according to elderly local informants the script had been created at Mando: Creswick, H.C., “On the Syllabic Character in Use Amongst the Vey Negroes,” Transactions of the Ethnological Society of London ns6(1868), 260–63.CrossRefGoogle Scholar Is it possible that when Jondu was destroyed, some survivors resettled at Mondo, and that this caused the two places to be confused in later traditions? See note 55 regarding Bukele's connection to Mando.

¹¹⁴ Cary/Carey, born a slave but later a distinguished African Baptist preacher and pastor in Virginia, having much to do with the foundation of the Richmond African Baptist Mission Society, in 1821 was joint leader of a party of emigrant Afro-Americans which left for West Africa and landed at Freetown. The previous year an initial party of emigrants under the auspices of the (white) American Colonization Society had settled on Sherbro Island. The location was largely due to the influence of John Kizell, a Freetown “Nova Scotian” of Sherbro origin who had developed trade between Freetown and the Sherbro district. A report of Kizell's appeal for missionaries had been noted by Cary, and Kizell, who had been associated with Afro-American emigration proposals since 1810, had worked with an exploratory ACS party in 1818. However, the 1820 Sherbro settlement rapidly collapsed and the survivors retreated to Freetown. Cary's party joined them and remained in Freetown until Cary and a number of other Afro-Americans left in 1822 for the newly-established settlement at Monrovia. There, for some years, the colonization interests of the ACS and the missionary interests of Cary clashed, and although by 1827 Cary was himself an administrator, it is likely that the transfer of the school to Vai country was, in part, Cary's attempt to “go out to the heathen.” He later explained, however, that whereas at one time it had been hoped to run both the Monrovia and the Big Town schools, having discovered that it would be impossible to fund both, “as soon as I succeeded in establishing the school at Cape Mount I gave up the school here and turned all the means towards the support of the school at Cape Mount.” (ABapFMS/Cary, undated letter to Baptist Board of 1828). For Cary's career see Fisher, , “Lott Cary,” 380–418Google Scholar; Fitts, Lott Carey Legacy, chapters 1–2; and for the Kizell and the Freetown connection, see Fisher, , “Lott Carey,” 387Google Scholar, and also Fyfe, Christopher, A History of Sierra Leone, (Oxford, 1962), 112–13, 132Google Scholar and Wilson, , Loyal Blacks, 219, 370.Google Scholar

¹¹⁵ Cary wrote: “I have made a visit to Grand Cape Mount. And while there, lost no time in endeavouring to determine what was the prospect of getting a school under way among them. They are very desirous that I shall establish a school there. I think if the board will support a school one year, after that time it may be conducted with very little expense; and all I am waiting for is books, and the opinion of the board on the subject. Please lose no time in getting books sent on for this object, for that is the largest field for labor on this part of the Coast. … If the Board of Missions intend to send a missionary to Africa, now is the time, and Grand Cape Mount is the place. I have the king's letter; and he has my promise for a teacher” (Fitts, , Lott Carey Legacy, 32–33Google Scholar). In June 1827 Cary reported that he had spent some seven or eight days at Grand Cape Mount “last season” (that is, before the rains), “on business of negotiation … but found it impossible to conclude any thing owing to the unsettled state of their government—the agent has since been up there but found the same difficulty. The Prince has not been able to erect a school house according to promise, as there has been a very great commotion among these natives during the year past.” He further referred to “the Mandingo priests who have been for years propagating their system of religion in that nation”’ with whom “a middling severe struggle” could be anticipated: ABM 7(1827), 304. A further letter from Cary of 20 December 1827 mentioned the conveyance of goods for the school from Monrovia to Cape Mount, including “spelling books” (perhaps significantly since such primers began with words presented in syllables), and described the school room, the arrangements for teaching, the palaver with “the king and all the headmen” which led to their consenting to the establishment of the school, and the opening day. Moreover, “as I had the king and his headmen present I got them to sign the articles of agreement in the presence of the whole congregation:” ABM 8 (1828), 144; Fisher, , “Lott Cary,” 406.Google Scholar Some of the spelling books were stolen en route, which Cary suggested was an encouraging feature. Cary's letters appeared in the American Baptist Magazine, but Fisher hints that Cary's writings may have been to some small extent “improved,” at least in English style, before appearing in print.

¹¹⁶ After the invention of the script, “a regular day-school” was formed in Jondu to teach it, but was later abandoned because of wars (Koelle, , Narrative, 24Google Scholar; Koelle, , Outlines, 238Google Scholar). This may have been in imitation of Islamic schools, where boys were taught to recite the Koran and read Arabic script.

¹¹⁷ As stated in note 113, Holsoe has suggested that “Big Town” represented the interior capital, Gohn Zolu Duma, of the local Vai ruler, Zolu Duma alias King Peter, since Revey reported the death of Zolu Duma from there. According to letters from Revey what exactly took place at “Big Town” in February 1828 was “the burial of old king Peter's bones” and a funeral wake lasting many days (ABapFMS/Cary, letters of Revey to Cary 13.02.1828 and 28.02.1828). Because the present town with the name Gohn Zolu Duma lies 35 miles inland, we are not convinced that the first school was set up so far from Cape Mount, the locality with which Monrovia had shipping connections, and at which there were Liberian contacts. It is possible that, after Zolu Duma captured the coastal districts of Gawula and Tombe in order to trade directly with the shippers (Holsoe, , “Cassava-Leaf People,” 123Google Scholar), he moved his capital and residence to a town within those districts, possibly Mando. This town was probably the site of the school (see note 113).

¹¹⁸ AR 4/2 (April 1828), 40.

¹¹⁹ At the opening of the school, Cary spoke “through brother John.” And perhaps John had some responsibility for the “king's letter” (note 115). Since John seems to have been in Freetown at the same time as Revey, and may have had the same religious contacts there, it is just possible that they already knew each other. The later history of John after the school closed is not known. John's first local convert, George, aged 24 or 25, being “determined to put away” one of his two wives, was said in 1827 to go “from house to house through the neighbourhood, reading and praying, and exhorting the people to repent and turn to God and have faith in Jesus Christ:” ABM 7(1827), 304.

¹²⁰ Cary wanted “to remove him from the persecution of the natives among whom he lived.” John gave as his reason for remaining that he had a wife and two children, but had not yet paid off the brideprice so was not allowed to leave, and Cary sent him money: AMB 7(1827), 303. In 1827 he wrote to Cary from Big Town to say that he would come to Monrovia, bringing with him a convert to be baptized there — later the convert, George, told Cary that he had “found God “ because “John learned him.” John said that he would travel “as soon as my ivory comes,” which seems to indicate that he was a trader. An Afro-American, “Brother Nelson,” wh o was probably also a trader, reported to Cary that he had encountered John Baptist at Big Town and been astonished by the sincerity of his Christian expression (“experimentally acquainted with the Lord Jesus Christ”): ABM 7(1827), 304.

¹²¹ HAM, 442.

¹²² AR 4/2 (April 1828), 40; repeated in MH 24/6 (June 1828), 188. The school was “superintended generally by the missionaries Carey and Lewis [the Reverend John Lewis, arrived 1826, later dismissed].” The Vai convert, John Baptist, assisted Revey at the school. Cary referred to “a native man belonging to and resident in the same town—who also I baptized a year or two ago. He has hitherto been very useful to the school. He can read so as to be of use to the children. He appears to have a great attachment to the school—and very sensibly feels for his country people and therefore lends every aid in his power to their improvement. I think he deserves encouragement. I have indevoured [sic] to make the establishment as profitable to him as the nature of our means would allow. He at present furnished the house for Brother Revey and cooking washing is done in his family — which brings in something but very little in comparison with what is necessary to keep himself and family above the native fashion. He served as an interpretor [sic] to Bro. Revey” (ABapFMS/Cary, undated letter of 1828).

¹²³ Peck, , “Missions,” 442.Google Scholar

¹²⁴ Revey came from New York, was “free-born,” literate, and aged 19 on departure (Shick, Tom W., “Emigrants to Liberia, 1820 to 1843: an Alphabetical Listing,” (1971), 25Google Scholar, the information being compiled from U.S., Congress, Senate, Roll of Emigrants that have been sent to the Colony of Liberia, West Africa, by the American Colonization Society and its auxiliaries, to September, 1843 & c., 28th Cong. 2d Session, 1844, V, vol. IXGoogle Scholar). For the 1820 emigrant party, see Coker, Daniel, Journal of Daniel Coker: A Descendant of Africa, From the Time of Leaving New York, in the Ship Elizabeth, Capt. Sebor, on a Voyage for Sherbro, in Africa, in Company with Three Agents and About Ninety Persons of Colour. The Rev. Samuel Bacon, John. B. Bankson, Samuel S. Crozer, Agents (Baltimore, 1820)Google Scholar; but Revey is not mentioned.

¹²⁵ Fisher, , “Lott Cary,” 405Google Scholar; ABM 7(1827), 305. Revey may well have been a teacher at a school on Sherbro Island in 1820–21, but after the settlers retreated to Freetown there is no record of a school surviving in that district, allegedly up to 1825. Presumably Revey moved to Monrovia in 1826 or 1827 and perhaps was a teacher there. Cary wrote that he had baptized the young man (Revey was now 26) and thought that he would be a blessing to the church. Further, “on proposing the subject to Brother John Revey, he is quite willing to go up [to Cape Mount] to commence the school as soon as the Brother comes down.” We have noted above the possibility that Duwalu Bukele was in contact with a “Poro” missionary at Sherbro. Could this have been Revey, and was this why Revey was agreeable to transferring his educational activities to Bukele's home town? The suggestion is, however, markedly conjectural.

¹²⁶ Fisher, , “Lott Cary,” 406Google Scholar; Fitts, , Lott Carey Legacy, 46Google Scholar, but both cite the account of the closure of the school in Peck, , “Missions,” 442Google Scholar, written in 1840. This actually reads “He [Revey] continued to take charge of it until the death of Mr Carey, when he was obliged, for want of helpers, to leave it.” Cary, who died unexpectedly in an accident, had forecast, however, that the initial funding would cover only twelve months. The earliest printed account of the closure, of June 1829, stated that “as Mr Carey had the entire control of the money sent out for that purpose, and until his affairs are settled, no disbursements can be made. It is already ascertained that the school at Grand Cape Mount is temporarily suspended until further remittances can be made:” ABM 9(1829), 195. In 1830 Revey claimed that he left the school “with much regret on both sides,” and ten years after the event, that it was “only want of support that induced him to abandon the field” (ABapFMS/Skinner, letter of 9.12.1830; Mylne, letter of–.07.1838). However, in a letter from Big Town three months after the school started, Revey complained bitterly that, during a tumultuous fortnight of wake for the deceased ruler, almost all the pupils had stayed away (ABapFMS/Cary, letter from Revey to Cary of 20.2.1828), suggesting that local enthusiasm and support was fragile and perhaps spasmodic.

¹²⁷ Peck, , “Missions,” 442.Google Scholar

¹²⁸ In 1829 it was reported from the U.S. home base that funds had been sent out to re-establish the Big Town school, and it was hoped that “the former instructer [sic], Mr John Revey has, ere this, renewed his labors:” ABM 9(1829), 195. In 1834 a report from Monrovia claimed that church members included “at Big Town near Cape Mount amongst the Veys two, one an exhorter, both natives—where brother Revey has taught a school for a considerable time, and numbers have learned to read:” ABM 15(1835), 114. The ambiguous wording of this report, while seemingly implying that Revey was not at Cape Mount in 1834, may be read as also implying that he had returned to the school for a time after its 1829 closure. However, we have traced no report between 1829 and 1834 which confirms that Revey did return. In December 1830 he was teaching at a Monrovia school, although there was still hope that the Big Town school “might be revived” (ABapFMS, correspondence of B.R. Skinner 1829–1831, letter to the Baptist Board of 9.12.1830). The April 1834 joint report of Wilson and Wynkoop does not mention any school at Cape Mount, but it refers to “a Baptist missionary among them [the Vey people at Grand Cape Mount] and the missionaries of the Wesleyan Missionary Society have determined to settle there also:” AR 10(August 1834), 182. This reference to a Baptist missionary is to A.W. Anderson (see note 104). According to Holsoe, quoting the Liberian Government Archives (not accessible to us), a Methodist Episcopal Church missionary, A.D. Williams, attempted to operate at Cape Mount in 1833 and 1834, but had to live with a slaver and withdrew when local wars occurred (Holsoe, , “Cassava-Leaf people,” 127, 29–30Google Scholar; cf. Foreign Missions of the Protestant Episcopal Church, 10 1860, 368Google Scholar). Williams seems also to have been at Cape Mount in 1830 but in what capacity is not clear: AR 6(1830), 211. During most of that year he was acting administrator of the Monrovia colony: Smith, James Wesley, Sojourners hi Search of Freedom (Lanham, 1987), 155.Google Scholar Williams, born in Petersburg, Virginia, had in 1823 traveled to Africa on the same ship as Curtis, and it is therefore likely that the “slaver” with whom he stayed at Cape Mount was Curtis (ibid., 37).

¹²⁹ In December 1835 Revey, “who formerly taught a native school at Cape Mount,” was living at Monrovia and had been “lately ordained as an Evangelist:” ABM 16(1836), 93, 133. In 1836 he “commenced a private school at Monrovia” and then became a superintendant of emigrants at Marshall on the Junk River (ABapFMS/Mylne, letter to Baptist Board of 29.11.1836; ABapFMS/Crocker, journal 23.9.1836). In 1837 Revey moved to Cape Palmas (Peck, , “Missions,” 442Google Scholar), where in 1838 he was met by the Bassa missionaries (ABapFMS/Mylne, letter to Baptist Board of -.07.1838). In 1840 Revey was still alive, since Peck spoke highly of him (“he commands the respect of all classes and denominations:” Peck, “Missions,” 443n).

¹³⁰ ”Mr Handt [later spelled Handz] has separated himself from society and gone to establish an independent mission at Cape Mount,” his intention being to succeed Revey, and he “is well and keeps school:” AR 5/1(March 1829), 25; cf. MH 25/10 (October 1829), 333. A colleague in the Liberia mission later reported that Handt “has come into the [Basel] Institution with a knapsack on his back, and so he must go out! Nothing could persuade him to stay and remain with us!” (CMSA, CA1/0196, report of J.F. Sessing, 1.5.1829). The Basel Mission had been encouraged to come to Liberia by Ashmun (Gurley, , Ashmun, 360Google Scholar), who was later criticized for attracting it by over-optimistic statements (cited in Du Plessis, J., The Evangelization of Pagan Africa (Cape Town, 1930), 98.Google Scholar For the arrival of Handt and a colleague at Sierra Leone on 10 December 1827 and the colleague's arrival at Monrovia on 21 December, see MH 24/6 (June 1828), 187. By June 1829 the Baptist mission authorities in the U.S., noting the temporary closure of the school, added that they had learned “with surprise … that a European Missionary had intruded himself upon their ground and sent home for the means of sustaining himself in the position:” ABM 9(1829) 195. Cf. “It [the school] was afterwards taught by a Swiss missionary” (Peck, , “Missions,” 442Google Scholar), “though the natives greatly preferred an English or American teacher:” ABM 9(1829), 213. That Handt went to Big Town but after “half a year” rejoined the other surviving Basel missionary, “exhausted,” is stated in Schlatter, W., Geschichte der Basler Mission 1815–1915, III Afrika (Basel, 1916), 12.Google Scholar A single long and informative letter (in German) from Handt at Big Town, addressed to the mission authorities in Switzerland, dated 20 December 1828, is preserved in the archives of the Basel Mission, but is not easy to decipher (Basel Mission Archives. Afrika Mission nach Liberia 1826–1832, 1829 Nr.3). It indicates that, debarred from a mission in Bassa country because of wars there, and acting on a list of suitable peoples for mission attention in Ashmun's invitation, Handt decided to go to Cape Mount, to resume Revey's school, and had been encouraged to do so by Cary's successor in Monrovia. He carried a letter from Revey to John Baptist, who presumably acted as his interpreter, and at Big Town also met George Peter. He had a session with the king and he described the school. But there is nothing in the letter to relate it to the invention of the script. The inventors are not named—he refers only to teaching children, not adults—and the form in which he was addressed locally (John?) is not noted. (We are much indebted to the present archivist, Paul Jenkins, for providing photocopies of the document; to Ulrike Sill for summarizing it and transcribing relevant extracts; and to Adam Jones for acting as intermediary and reading and checking items of information.) After returning to Monrovia, Handt traveled to Sierra Leone and joined the Church Missionary Society, which he later served for several decades in another mission field, Australia.

¹³¹ It might, however, be argued that the conclusion does not necessarily follow. Perhaps Bukele and Revey (if he was John), being both non-white, yet with experience of “whiteman's culture,” were appraising themselves over against the young men who lacked the experience. Hence the young men's sharp reaction.

¹³² “Individuals who have English names have them generally unconnected with, and independent of, their ‘Country-names’, and use them almost exclusively with English-speaking people” (Koelle, , Outlines, 20Google Scholar).

¹³³ It is worth noting, however, that an exaggerated claim of illiteracy may be advanced in order to accentuate the miraculous or sensational aspect of subsequent proceedings, as with the dictation of the Koran by Muhammad, and more relevantly, the invention of the Cherokee syllabary by Sequoyah.

¹³⁴ Forty-first Annual Report of the American Colonization Society (19 01 1858), 16.Google Scholar The relevant passage dealing with the invention, written by an anonymous author, comments that “there is reason to think [it] originated in the teachings of a Liberian missionary.” The passage then cites (from “the first report of the Trustees of Donations for education in Liberia,” no date given) a statement about the invention by the Rev. Joseph Tracy (who in 1840 had written the contribution on the missions of the ABCFM in the History of American Missions). Tracy had stated that Bukele had learned the Roman alphabet from an American missionary (an interpretation of Koelle's account). The anonymous author continued: “The missionary was doubtless a Liberian, the Rev. John Revy, who taught a school for heathen children at Cape Mount about 1825. He was afterwards well known, as a Baptist preacher, and Colonial Secretary at Cape Palmas.” The sentences quoted in our text appear in a footnote in the original, probably contributed by the anonymous author. The attribution of influence on the invention to Revey appears to be mere supposition. No source or reference is cited; the date for the school is wrong, as may be the statement that Bukele had still to learn the Roman alphabet, while the “doubtless” does not indicate firm information. The reference to a dream almost certainly came from the commentator having read Koelle's account, published a few years earlier. Revey and his school were mentioned in the 1840 History of American Missions, when Revey was apparently still alive and active, but no connection between him and the invention was then suggested. Had the 1858 claim of a connection come from Revey himself, it might be either suspect or genuine, but the report reference reads as if by then Revey was dead. The author of the report's knowledge about Revey's school was perhaps obtained from the History of American Missions, and his suggestion that “a Liberian missionary,” otherwise Revey, may have had some connection with the invention may have been no more than a guess. In any case, it would seem that he was claiming at the most a very indirect missionary influence since, in the previous paragraph, he had quoted from Wilson's book of 1856 the view that the Vai inventors did not, “it is believed, receive any assistance, whatever, from any one, in perfecting this wonderful invention.”

¹³⁵ Conflating Koelle, , Narrative, 23Google Scholar; Koelle, , Outlines, 237.Google Scholar

¹³⁶ Koelle, , Narrative, 23.Google Scholar

¹³⁷ Koelle estimated Bukele's age to be about 40 in 1849. If so, he would have been 23 or 24 at the time of the invention, and about 18 in 1827–28, the years in which Revey labored in Vai country, so hardly “a little boy.” A youth of this age would probably not have had much influence or respect in a West African society—Koelle wrote that “all of these six men were then still young, being all from twenty to thirty years of age. They were therefore afraid people might not pay them proper attention” (Koelle, , Narrative, 24Google Scholar). It is therefore possible that Bukele was older than Koelle thought, and certainly his fellow inventors seem to have regarded him as their senior. (For what it is worth, in 1834 Wynkoop reported that the individual who “dreamed that he must immediately begin to make characters for his language” was an “old man.”) Alternatively, it is just possible that mature Vai used the term “little boy” to refer dismissively to an unmarried youth.

¹³⁸ It is arguable that Bukele is more likely to have remembered verses of the Bible (in English) from an experience of a few years earlier than from one several decades earlier.

¹³⁹ The earliest documentation of the Vai script in a printed Baptist source appears to have been in 1840. No reference to the 1834 report of the script by Wilson and Wynkoop was noted: “The Veys are superior to most native tribes. They have made some progress toward civilisation, and are said to have a written language, of their own invention, by which they communicate with each other, although they have no books” (Peck, , “Missions,” 441Google Scholar). This report contradicts the published 1834 report by Wilson and Wynkoop that the Vai had books in the script. The source of Peck's information was almost certainly the Bassa missionaries. Crocker had corresponded with Peck in the later 1830s, and Mylne in 1838, referred to the “Veys” who, “I am told, have a method of writing invented by themselves, by means of which they communicate with each other; but whether they have written any books, I have not been informed,” Mylne having been “told” almost certainly by Revey whom he mentions in the same letter as having recently been met at Cape Palmas and with whom he had had a “good deal of conversation” (ABapFMS/Mylne, letter to Baptist Board of–.07.1838). Peck, although referring to the Vai script and although discussing at some length Crocker's activities, made no mention of the latter's experiment with a Bassa syllabary, presumably because it had been abandoned.

¹⁴⁰ ABapFMS/Crocker, letter to Baptist Board of 28.10.1835; ABM 16(1836), 93. Mylne added that Revey “is said to be very successful in communicating instruction to the natives. He was lately ordained as an Evangelist, by the First Baptist Church in Monrovia and has good report as an excellent, judicious brother.”

¹⁴¹ On Pinney, see notes 34 and 102.

¹⁴² ABM 16(1836) 196-97.

¹⁴³ ABapFMS/Crocker, journal 18.4.1836.

¹⁴⁴ By September 1836 Revey was “employed as a superintendant of emigrants at Marshall on the Junk River” (ABapFMS/Crocker, journal 23.9.1836). Conceivably Crocker and Revey met while the latter was at Junk River but by September 1836 Crocker had invented, used, and abandoned the Bassa syllabary.

¹⁴⁵ ABapFMS/Mylne, letter to Baptist Board, July 1838.

¹⁴⁶ Of course it is plausible that both texts contain errors because of their mode of collection and that these, if known, would explain some of the difficulties.

¹⁴⁷ Moreover, Bukele's missionary left the country after three months, whereas Revey left Vai country after about a year.

¹⁴⁸ Yet in 1836, at a Liberian school a missionary was “teaching the alphabet to a class of nine persons, from the age of 10 to 40:” AR 12(September 1836), 280.

¹⁴⁹ The statement of Wilson in 1856 is not entirely free from ambiguity. He insisted that “the characters used in this system are all new, and were invented by the people themselves,” a point not in question. But his following assertion that the Vai did not “it is believed, receive any assistance whatever from any one in perfecting their wonderful system,” can be understood as less than a categorical denial of external influence by the reservation of “it is believed” and by the use of “perfecting” rather than “creating.” Even “from any one” sounds defensive, and surely relates to an earlier moralizing claim that the invention was “enough to stifle the cavils and sneers of those who think so contemptuously of the intellectual endowments of the African Race” (Wilson, , Western Africa, 95Google Scholar). It can reasonably be accepted that Wilson was genuinely unaware of any connection between the Cherokee and Vai scripts, yet was not disposed to go so far as to claim that the syllabic mode of the characters was indisputably without external influence, for which see the next note.

¹⁵⁰ In 1856 Wilson wrote that “the idea of communicating thoughts in writing was probably suggested by the use of Arabic among the Mandingoes, and from the practice of whites who occasionally visited this country for the purpose of trade” (Wilson, , Western Africa, 95Google Scholar). This was an involved missionary specifically ignoring, indeed almost implicitly excluding, any missionary influence on the invention. But since the context was one in which the author felt it necessary to speak in glowing terms of the invention of the script because it confounded those inclined to scorn “the intellectual endowments of the African race,” he perhaps thought a strategy of missionary modesty morally justified.

¹⁵¹ Hair, , “Notes,” 40.Google Scholar For Wilson (1809–86), see Anderson, Gerald H., ed., Biographical Dictionary of Christian Missionaries, (New York, 1998)Google Scholar; Wynkoop (1806–70), after his visit to the coast in 1834, completed his theological training and wrote on the slave trade in Liberia, but appears never to have returned to Africa.

¹⁵² Cape Mount was the only district on the Liberian coast where the missionaries confronted a dominant and aggressive Islamic influence, and therefore they tended to concentrate on the other districts, where their task was easier and more successful.

¹⁵³ See note 40 for an 1836 passing reference by a missionary to the use of the script, without the missionary making any comment on the circumstances of the invention.

¹⁵⁴ Prior to removal, the Cherokee occupied parts of what are today Virginia, Georgia, Alabama, Tennessee, Kentucky, South Carolina, and North Carolina.

¹⁵⁵ Shick, , “Emigrants,” 25.Google Scholar For information on the Oswego, see Brooks, George, Yankee Traders, Old Coasters and African Middlemen (Boston, 1970), 311.Google Scholar At dates between 1823 and 1833 other individuals with the name Curtis, men and women who also came from Virginia, arrived in Liberia—were they his relatives?

¹⁵⁶ Holsoe, , “Cassava-Leaf People,” 51.Google Scholar

¹⁵⁷ Curtis would later turn on Canot: Mayer, Brantz, Captain Canot; or, Twenty Years of an African Slaver Being an Account of his Career and Adventures on the Coast, in the Interior, on Shipboard, and in the West Indies (New York, 1854), 444–47.Google Scholar

¹⁵⁸ Holsoe, , “Cassava-Leaf People,” 127–28.Google Scholar

¹⁵⁹ Cain was a war leader from Gonamalo (Gnama) (ibid., 134).

¹⁶⁰ Mayer, , Canot, 444.Google Scholar

¹⁶¹ American Anti-Slavery Society, 1834, 12.

¹⁶² Mayer, , Canot, 444.Google Scholar

¹⁶³ Forbes, Frederick E., Six Months' Service in the African Blockade (London, 1849), 70.Google Scholar This reference by Forbes is the only known reference to Curtis' Cherokee background. Holsoe stated that Curtis was “half Cherokee,” and this was repeated by subsequent writers (Holsoe, , “Stimulus Diffusion,” 51Google Scholar; Dalby, , “Historical Problem,” 119Google Scholar). We have seen no evidence to corroborate this, although it may have been the case that Curtis was of mixed Cherokee and Afro-American parentage, providing an explanation for his immigration to Liberia. For information on the African Benevolent Society (an association founded in 1829 by the slaves of Cherokee planters for the support of Liberia) and immigration to Liberia by free blacks from the Cherokee Nation, see Perdue, Theda, Slavery and the Evolution of Cherokee Society, 1540–1866 (Knoxville, 1979), 92–94Google Scholar; and Halliburton, R. Jr., Red over Black: Black Slavery Among the Cherokee Indians (Westport, 1977), 99.Google Scholar For a note on Cherokee support for the American Colonization Society in missionary correspondence and in the Cherokee Phoenix of 1828, see McLoughlin, William, Cherokees and Missionaries, 1789–1839 (New Haven, 1984), 130.Google Scholar

¹⁶⁴ Forbes, , Six Months' Service, 70.Google Scholar

¹⁶⁵ Ibid., 70–71. A letter in English from Curtis to Forbes refers to a local upheaval and a robbery at his house, allegedly to seize and destroy a recent anti-slave trade treaty, with the loss of “all your washed clothes.”

¹⁶⁶ “Commander Forbes, R.N., and Liberian slave trade,” AR 28/2(02 1852), 58.Google Scholar For Curtis' role in the early 1830s, see note 5. In the 1840s there was some belief that the British, advancing downcoast from Sierra Leone might be interested not only in claiming Gallinas, another Vai territory, but even challenging Liberian claims to Cape Mount.

¹⁶⁷ Koelle, , Narrative, 7–8.Google Scholar Probably “Beach Town” and “Sandbeach” were the same place. For the possibility that Curtis was of mixed Afro-American and Cherokee ancestry, see note 163. Koelle failed to name his host, but this may have been because of his disapproval of the “concubines.” However, if this was indeed Curtis, and if in fact Curtis was the individual who had suggested a syllabary to the Vai, it is strange that Koelle, who was investigating the syllabary, did not learn from him about the connection, and then mention it. It seems more likely that either this was not Curtis, or Curtis was not the individual who had made the suggestion.

¹⁶⁸ Forbes, , “Despatch,” 90Google Scholar; the same report by Forbes, in his book, Dahomey and the Dahomans, (2 vols.: London, 1851), 1:199.CrossRefGoogle Scholar Note that Forbes considered Curtis “one of the Liberian settlers,” although arguably he was an independent trader working at times in association with well-known slave traders. By “language” Forbes seemingly meant “script.” He explained that he could not have the characters interpreted until he met elsewhere a Vai literate in the script, thus implying that he had copied the inscription. He may well not have copied it exactly, and when he heard the man sounding out the Vai syllables may not have heard them exactly. The name of “Lieut. Forbes' interpreter” was supplied and discussed by Koelle (Koelle, , Outlines, 20Google Scholar).

¹⁶⁹ Forbes, , “Despatch,” 110.Google Scholar Koelle wrote: “I have myself seen the few indistinct characters, written with charcoal on the walls of a house, which had first attracted his [Forbes, ] notice [to the script]” (Koelle, , Outlines, viGoogle Scholar). This implies that Koelle was unable to check on Forbes' transcript of the characters. It may be noted that Koelle did not mention the inscription in his 1849 account (although he may have lived in Curtis's house).

¹⁷⁰ Welmers, William, A Grammar of Vai (Berkeley, 1976), 3.Google Scholar Gail Stewart, a missionary educationist and linguist, died in 1998. Hair is grateful for having had the opportunity to work with Mrs Stewart on Vai script projects in the late 1960s and early 1970s, after both had previously lived and worked in West Africa. They had hoped to make an overall assessment of changes in the shapes of the characters between the time of the invention and the present day (as noted in Dalby, “Indigenous Scripts”, 16n1), but after collecting a quantity of material, the task exceeded Hair's grasp and other commitments intervened. Mrs Stewart, however, presented an important conference paper on the original form of the characters (Stewart, “Early Vai Script”).

¹⁷¹ Ibid., 20. Koelle, who had seen the characters and probably met Curtis while at Cape Mount, in his Vai grammar includes, in a list of Vai words adopted from English, “Curtis, a proper name V. Koísi” (Koelle, , Outlines, 12Google Scholar). He may have collected the Vai term and what it stood for while in Vai country, or he may have later taken the term from Forbes' publications and deduced the gloss.

¹⁷² Curtis was almost certainly considered by the Vai a “Poro,” translated by Koelle as a “white man,” although fully or partly a Cherokee, and just possibly part Cherokee and part “Black” in ancestry.

¹⁷³ Was Curtis sufficiently politically conscious to warn the Vai of the dangers of encroachment by “the civilized,” equating the past Cherokee experience of the American whites with the likely future Vai experience of their Afro-American neighbors?

¹⁷⁴ See note 27.