The name 'Akan'
When you learn Akan - whose language do you learn?
The word 'Akan' designates quite different groups of people depending on the period of time at which it was used and on the context in which it was or is being used. Roughly, we can distinguish between its traditional native use, its use as a scientific classificatory term, and its (modern) socio-political use.
In Christaller's understanding, the term 'Akan' covered the 'Akan-speaking' people of Ghana (the then Gold Coast), not including the Fante in the South. The term 'Twi' ('Tshi'/'Chwee' [Christaller 1875, 1881] are old notational variants not currently used) would have served, according to Christaller's proposal, to denote the same group of people and language varieties, but would also have included the Fante. Christaller recognised that the Fante were closely related to the other groups. His inclusive use of the term 'Twi' served to stress the fundamental linguistic unity between those groups and the Fante. Thus in Christaller's terminology, 'Twi' is the more inclusive ethnoglossonym, covering also the Fante, whereas 'Akan' (which he rarely uses) is the more restrictive term, excluding the Fante.
In order to avoid confusion, it is
important to note that current usage of these two terms,
Akan, and Twi, is the reverse from Christaller's. In modern
usage, Akan is the more inclusive term, which covers the
whole of the Akan-speaking people of Ghana, whereas Twi is
the less inclusive term, being conventionally used to
identify a subgroup of Akan dialects and their speakers. As
it is used by the people themselves, as well as in the
modern literature, the Twi comprise notably the Akuapem and
the Ashanti (or 'Asante') but not the Fante.
On the other hand, the wider designation 'Akan', according to modern usage, includes the Fante. A number of other groups such as the Brong and the Wasa are also subsumed under the general designation of Akan (as they were already at Christaller's time) but do not consider themselves to be Twi. Thus the current use of the term Twi is more limited than what was covered by the original term 'Akan' as explained in Christaller's dictionary.
The term 'Akan' is also used as a name for self-identification by a number of populations extending from what is recognised as present-day Akan territory towards the west including large parts of Eastern and Central Ivory Coast. From a linguistic point of view, all these groups are indeed closely related. In the scientific literature, they used therefore to be classified as Akan, e.g. in Greenberg (1966) -> Annex 1: General bibliography. In this terminology, which was, and sometimes still is, used by anthropologists and linguists, the term 'Akan' includes, beyond the Akan populations of Ghana, the Baule and Anyi of Ivory Coast. From a primarily linguistic viewpoint, one could consider that the Nzema of Southwestern Ghana are also part of this dialect-language continuum. The Chakosi, who now live in the northern border area of Ghana/Togo, isolated from the rest of Akan, are also linguistically part of it (-> geographic distribution, tree diagrams 1 and 2).
As part of an emerging national language policy in the 1950-ies, the present restricted use of the term 'Akan' was deliberately promoted as the official designation of the Akan-speaking community of Ghana (Dolphyne 1986, 1988). In order to keep this official usage distinct from the wider scientific and classificatory term designating all the groups speaking languages related to Akan, the term 'Tano' or 'Potou-Tano' was proposed by Stewart. These terms were coined in order to replace 'Akan' for denoting the larger group of Akan-related languages. According to this more recent terminology, Akan as a scientific and classificatory term takes its place at a lower level and is limited to the cluster of dialects which corresponds to the official Ghanaian designation, as can be seen in the following diagram.
While Christaller was doubtlessly right in stressing the linguistic unity of all Akan peoples, there were historical reasons both internal and external which led others in his time to stress the differences rather than the unity. On objective, linguistic grounds, however, the unity of Akan cannot be doubted.