Ghana appears to have had a central…

Ghana appears to have had a central core region and was surrounded by vassal states.

One of the earliest sources to describe Ghana, al-Ya'qubi, writing in 889/90 (276 AH) says that "under his authority are a number of kings" which included Sama and 'Am (?)

and so extended at least to the Niger valley.

These "kings" were presumably the rulers of the territorial units often called kafu in Mandinka.

The Arabic sources, the only ones to give us any information are sufficiently vague as to how the country was governed, that we can say very little.

Al-Bakri, far and away the most detailed on, does mention that the king had officials (mazalim) who surrounded his throne when he gave justice, and these included the sons of the "kings of his country" which we must assume are the same kings that al-Ya'qubi mentioned in his account of nearly two hundred years earlier.

Al-Bakri's detailed geography of the region shows that in his day, or 1067/1068, Ghana was surrounded by independent kingdoms, and Sila, one of them located on the Senegal River was "almost a match for the king of Ghana."

Sama is the only such entity mentioned as a province, as it was in al-Ya'qubi's day.

In al-Bakri's time, the rulers of Ghana had begun to incorporate more Muslims into government, including the treasurer, his interpreter and "the majority of his officials."

Given the scattered nature of the Arabic sources and the ambiguity of the existing archaeological record, it is difficult to determine when and how Ghana declined and fell.

The earliest descriptions of the Empire are vague as to its maximum extent, though according to al-Bakri, Ghana had forced Awdaghast in the desert to accept its rule sometime between 970 and 1054.

By al-Bakri's own time, however, it was surrounded by powerful kingdoms, such as Sila.

Ghana will be combined in the kingdom of Mali in 1240 marking the end of the Ghana Empire A tradition in historiography maintains that Ghana fell when it was sacked by the Almoravid movement in 1076–77, although Ghanaians resisted attack for a decade.

but this interpretation has been questioned.

Conrad and Fisher (1982) argued that the notion of any Almoravid military conquest at its core is merely perpetuated folklore, derived from a misinterpretation or naive reliance on Arabic sources.

Dierke Lange agrees but argues that this doesn't preclude Almoravid political agitation, claiming that Ghana's demise owed much to the latter.

Furthermore, the archaeology of ancient Ghana simply does not show the signs of rapid change and destruction that would be associated with any Almoravid-era military conquests.

While there is no clear cut account of a sack of Ghana in the contemporary sources, the country certainly had converted to Islam, for al-Idrisi, whose account was written in 1154, has the country fully Muslim by that date.