Toponyms: Neglected wallflower or pot of plenty

Threats of imminent extinction motivate language documentation; they also allow place name neglect. This paper examines settlement names within Africa’s Edoid group. Village nomenclature converges on a restricted range of conventions; however, interethnic contact has led to non-Edoid toponyms for three villages. Two derive from the trade language Hausa. A third links to Igbo blacksmiths supporting rainforest penetration with iron tools, as is evident in cognate vocabulary. Iron use most naturally follows a pastoral era outside the rainforest, which number prefixes on herd-animal nouns support. Toponymic studies thus remind us of the benefit accrued when documentation looks beyond “the single ancestral code.”


Introduction.
The threat of imminent extinction has motivated language documentation studies but also allowed for neglect of place names.Indeed, recent linguistic study has emphasized toponyms as fixed in space (Sims Williams 2006) and relics from deep historical time.Too easily they become, like other highly specialized terms, neglected wallflowers of fieldwork (Evans 2010).The stability and persistence of place names are amply illustrated for many areas of the world (Burenhult andLevinson 2008, Senft 200, Nash 2013).But these recent and stimulating findings have tended to ignore sub-Saharan Africa, especially its many languages (Heine and Nurse 2000) and vigorous multilingualism (Vigouroux and Mufwene 2008).A case in point concerns the Benin Kingdom of precolonial West Africa's forest zone (Ryder 1969, Connah 1975) and its Edoid languages.
The Edoid languages are spoken in south-central Nigeria.From them emerged the Benin Kingdom of pre-colonial West Africa (Bradbury 1957(Bradbury , 1973)).Our understanding of this kingdom derives in large measure from oral history (Egharevba 1934), the highly stylized bronze casting and ivory carving of the 14th through 17th centuries (Crowder andAbdullahi 1979, Ben Amos 1980) and archaeological investigation of its capital, Benin City, where earthen embankments and palace grounds have been uncovered (Connah 1975, Darling 1998).These assessments point to the Kingdom's patrilineal, hereditary kingship and centralized palace government (Ogbomo 1997) The three clan cluster consisting of Emai-Ora-Iuleha (EOI) occupies some twenty two villages: 10 Emai, 6 Ora and 6 Iuleha.The names for these villages are indicated below under their respective clan heading.Overall, village nomenclature converges on a restricted range of naming conventions.However, interethnic contact in this endangerment hotspot (Romaine 1973, Nettle andRomaine 2000) has led to some non-Edoid toponyms.It is these outlier village names that invite us to revisit their fit.Before proceeding further, we should take note of where in Nigeria these three Edoid clans are located.In the south-central region, they occupy geographic areas in the transition zone where rain forest and savanna intermix.Class three names are morphologically bilingual and binomial, i.e. split-names.Each employs an Edoid name in construction with English phrase New-Site.

Emai:
Evbiame-New-Site Ojavun-New-Site Ora: Oke-New-Site Class four names are morphologically monomorphemic, person's name, bi-morphemic with name and modifier, or are unknown.

Uzebba Sabongida Ogute
Two of these reflect a double-naming strategy assigned to senior villages.
The source for these names is Hausa, spoken north of the EOI savanna/rainforest transition zone and the Niger-Benue confluence.Hausa has been employed extensively in commercial trade reaching EOI during at least the 19 th century (Bradbury 1957).Although these names reflect relatively recent contact, they have super ceded the aboriginal ones in utility, as roadway signposts attest.
A third "other" name is Emai's Ogute.Its source is Igbo, which, farther east toward the Niger River, abuts Edoid but not EOI.As a family name, Ogute is tied by elders to Igbo blacksmiths.Since cognates for iron artifacts also occur, e.g.ópìà 'cutlass,' it is likely that for centuries Igbo smiths provided tools necessary for maize and yam farming and, indeed, rainforest penetration.
A more complete analysis of EOI village names can be found in Appendix 1.

Figure 2 .
Figure 2. Map of EOI key administrative villages (Afuze, Sabongida-Ora and Uzebba) in their transition zone context of grassland and rainforest vegetation.

. Their nearest neighbors are the Yoruba to the west and the Igbo to
Village names across EOI clans exploit five morpho-semantic design principles.Their constituents (often phonologically diminished) reflect five source classes: (Childs, Good and Mitchell 2014)-semantic properties of EOI village names suggests at least a three-stage history of contact.Coding of grammatical number is asymmetric relative to economic lifestyle, as suggested by data from Emai.There are 11 remnant noun classes with contrasting singular/ plural nominal prefixes.Few if any lexemes strongly associated with agriculture exhibit a number prefix.Neither staples of the Emai diet nor tools for farming show prefixes.Shared lexical items further strengthen the nature of contact in EOI prehistory.Lexemes shared with Igbo populations have a distinct semantic character: iron tools, farming, animals, rainforest illness and social order.I conclude that EOI village names combine with other linguistic data (shared vocabulary and alternating grammatical number prefixes) to favor a five-stage prehistory for EOI and Edoid: herders outside the rainforest followed by rainforest penetration as farmers with iron tools and then British colonial occupation.Most importantly, though, EOI toponymic studies not only remove a wallflower but also identify the benefit of undertaking documentation beyond "the single ancestral code"(Childs, Good and Mitchell 2014).