I've read that national borders drawn during decolonisation caused long-term instability, especially in Africa and western Asia. How could the borders have been drawn differently to create a better outcome?

To start off: I can't really speak to the situation in western Asia because I don't feel knowledgeable about that region to speak competently.

When it comes to the continent of Africa, folks usually point to instances of ethnic violence (like the violence after 2007 elections in Kenya) or secessionist efforts (like Katanga, Biafra, Eritrea, Cabinda) and diagnose the problem as "bad borders". That is to say, state borders drawn during the colonial era that both (1) divide ethnic groups into multiple different states, and (2) place ethnic groups that had no pre-colonial history of living in the same kingdom/state together, causing conflict.

These critiques of the "artificial borders" of African states rely on a logic of the Nation-State. These critiques argue that culturally similar nations like the baKongo, Somalis, or Turaregs that are currently split across borders would be much happier living in one nation-state.

They also point to instances like Biafra secession, South Sudan secession from Sudan, Lozi secessionism, or the Anglophone crisis in Cameroon as instances where nationalist sentiment exists among marginalized and minoritized communities within a mutli-ethnic state.

So, if borders could be redrawn, or had been drawn more intelligently and more fairly at the outset these countries would have much fewer problems.

I would reply to these critiques by saying it's not that simple.

First off, in the pre-colonial era it was not "natural" for nationalities or ethnic groups to be governed as one single state. For instance, the Yoruba people of West Africa had numerous states like Oyo, Ile Ife, Popo, Owo, etc which were distinct and self-governing. When the British arrived in West Africa from 1830s to 1890s, they made treaties and alliances with certain of these states, while the French armies invading the Dahomey kingdom made treaties with Dahomeyan subjects who are now considered to be ethnically Yoruba.

So, it is not that the British and French conspired in this case to spit up the Yoruba nation. In the 1880s or 1890s, it would be anachronistic to speak of a singular Yoruba nation that could be "split up", but rather it was this array of Oyo, Ilesa, Ondo, Akure and other communities which saw themselves as distinct from each other. However, as colonialism brought these groups into the colonial Nigerian state and into contact with much different peoples like Hausa, Fulani, Warri, Tiv, and Igbo; these distinct communities began to change attitudes and come to accept a greater Yoruba national consciousness.

Now, on the other hand, we can point to other examples like in the horn of Africa where colonial powers like Britain and Italy engaged in diplomatic horse-trading, and how the British ceded the eastern half of Jubaland to Italian Somaliland in 1926 in reward for Italian participation in the First World War.

So, in some cases there truly was very artificial creation of borders that split up historical regions without consulting the resident populations.

Another thing to consider is that there was a great deal of movement of peoples during the near-century of colonial rule.

For instance, the French used the Upper Volta (now Burkina Faso) as a sort of labor reserve area for Cote d'Ivoire, with large numbers of migrant laborers moving to Cote d'Ivoire.

Ditto, there was lots of movement in Belgian Congo, Northern Rhodesia, Nyasaland, Bechuanaland, Southern Rhodesia, and South Africa of migrant laborers to the mines of the Copper Belt on Katanga/North Rhodesia border and the Gold and Diamond belts in northern South Africa.

Ditto, in Belgian Congo, there was the large-scale migration of people pushed by the expanding Rwanda kingdom, and then later in the 1880s-1920s settlement was encouraged by labor demands of colonial authorites in eastern Belgian Congo in the Kivu regions.

In East Africa and South Africa there was substantial in-migration of South Asian communities, first as contract labor and then as these communities became established they began to occupy a market niche as merchants and small business owners in cities and in the countryside.

Ditto, there was the emergence of Lebanese diaspora in French West Africa and in western Belgian Congo in the 1900s, who also occupied a place as merchants and business owners.

Finally, the colonial period was a nascent period of urbanization, with the flow of peoples from the countryside into cities. Almost by definition, that meant that different ethnic groups from across the colony would exist side-by-side in the major cities like Kinshasa, Nairobi, Johannesburg, or Algiers.

So, realistically, I think it would have been extraordinarily difficult to have re-drawn the borders of Africa during the late-colonial era or in the immediate post-independence era. Even secessionist states like Biafra and Katanga represented a predominant ethnic group(s) (Igbo and Luba/Lunda respectively) but these secessionist states also included numerous minority ethnic groups which were not all necessarily pleased to leave Nigeria or Congo and exist in a new state under Igbo or Luba/Lunda hegemony.

Finally, I'd argue that post-colonial African states have attempted methods to reduce ethnic tension. And also that ethnic tension arises because of political culture and decisions of political leaders within a country, not because different ethnic groups have "ancient hatreds".

To the first point: in the 1960s, 1970s and the early 1980s, sub-saharan African states frequently had relatively large and expensive public sectors (i.e. government, police, army, postal service, etc). Public sector payroll plus economic development programs were quite expensive, outstripping African economies ability to pay directly through taxation, so African states tended to rely on international loans, which international financial experts argued was responsible and affordable at the time.

However, with the world economic shocks of the first and second oil crisis in 1973 and 1979-80 (respectively), African states almost all faced economic crises and debt crises as the inflation rate for international loans rapidly increased.

These African states were forced by desperation to accept loan programs from the IMF and World Bank which came with strings attached, one of which was a requirement for layoffs in public sector to "sustainable" payroll levels.

The problem with this approach is, these public sector jobs had been a way that African presidents had doled out jobs and influence throughout the different ethnic groups within a country, and avoided charges of neglect and ethnic discontent.

So, with much reduced staff and payroll, some economic scholars like Ruth Rempel have argued that secessionism and ethic discontent increased during the "structural adjustment era" of the 1980s and early 1990s.

Also, recent polling (from 2005-2010) has argued that ethnic tensions within Sub-Saharan African countries tend to be higher in election years than in non-election years. The Kenyan writer Ngugi Wa Thiongo noted that tensions after the 2007 elections between were between Gikuyu and Kalenjin communities. These populations had homelands that did not border each other, and there had not been a history of Gikuyu-Kalenjin conflict in the pre-colonial or colonial era. Ngugi wa Thiongo argued that the conflict in 2007 between the two largest ethnic communities in Kenya represented a strictly political battle between politicians and communities who saw the stakes as control over government and economic benefits of control over government.

I think that this is actually a widespread phenomenon in post-colonial Africa, where the stakes are really economic and political competition over scarce resources.

And there are some instances where this political competition has played out in ways that confound the ethnic narrative.

For instance, in 1958 and 1959, there was political mobilization and violence coming from the Luba community of Katanga against migrant laborers coming from neighboring Kasai province to work in the copper mines of Katanga province. The thing is, the migrants were from the Luba community of Kasai, and were being attacked and forced out by the Luba of Katanga who were working with other "native katangan" communities.

TL;DR- focusing on borders as a source of ethnic discontent is not the whole story. Rather, it is political cultures (legacies of the colonial era) that encourage competition over economic and political power based on ethnicity; as well as economic crises, which have contributed greatly to instability.