Many sections of Abu Zakariyya al-Azdi’s Tarikh al-Mawsil show the significance of kinship ties and tribal identity in northern Iraq during the early 9th century. A common interpretation of this is that, with the weakening of the state during and after the Fourth Fitna, political contests were increasingly fought at the local level, with the side often apparently determined by competing tribal identities which had carried over from at least the garrison towns of southern Iraq. Carefully noting the patterns of different kinship networks over time within the general political, social, and economic environment, however, reveals that the caliphate had been and continued to be a crucial element shaping the patterns of power and influence within and among kinship networks even as the tribesmen adapted to that element based on their own interests. This paper will focus on three episodes and show the ways in which tribalism in and around early 9th century Mosul was influenced by the caliphate. One of these is the feud between al-Azd and Hamdan, which led to the latter’s exile from Mosul. Evidence from both al-Azdi and Ibn Hawqal indicates that caliphal land grants and formal state offices were crucial to the development of that conflict. The second is the relationship between al-Sayyid b. Anas, a Mosul office-holder, and his bother-in-law, Zurayq b. Ali. The pattern of their cooperation and subsequent conflict highlights the ways in which kinship networks co-existed with the state, leading to distinct forms of state power both within urban Mosul and in mountainous areas beyond its formal domain. The third is the account of a prominent Mosul family whose roots are traced to pre-Islamic Arabia. Viewed alongside parallel accounts in other sources, this narrative shows the manner in which the pre-Islamic Arabian past was marshaled to serve the status claims of an urban Islamic elite, as well as the ways in which that past was adapted to serve that end. Beyond the discussion of Mosul itself, this is intended to make two larger points. The first is that current models of empire can prove useful in explaining the Abbasid caliphate and the durability of Arabo-Islamic culture. The second is that, as social scientists increasingly argue, the state is a social relation, the actual power of which is frequently distributed among a range of networked elites.