There is a new rebel-led government in Bangui after Seleka rebels marched on the capital city just over a month ago, deposing the beleagured government of Francois Bozize after several months of shuttling between guerilla war and negotiations. This brought an upright end to the Libreville Accord brokered by the Economic Community of Central African States (ECCAS) in an attempt to avoid a military take-over, considered unconstitutional by the region and the African Union, from taking place and plunging the region of fragile states into an abyss of crises.

The Accord had forced Bozize to reach out to the opposition forces and include them in his government in order to increase the legitimacy of the state and the participation of marginalised actors. It had conformed to the practice of consoctional politics in Africa where governments of national unity are created, often via negotiations rather than elections, to give space to disgruntled and armed groups, in a desperate attempt to avert conflict. Often this distorts civilian governments and can legitimize rebels.

Now, South Africa became involved due to a 2007 bilateral agreement on military cooperation, which made provision for it to provide training to the CAR security forces and VIP protection support to the Bozize government. When the agreement expired in December 2012, the two governments renewed it even though we already knew about many questions being raised about the legitimacy and stability of the Bozize government. We, therefore, knew that the new agreement period would become complex and fragile.

To be fair to the Zuma government, contrary to media and opposition party criticism, the government communicated this through normal channels of communication including a DIRCO press statement in December 2012. In that communication it also indicated that this was a renewal in the belief that democratically-elected governments in fragile situations needed to be enabled to exercise authority over their territories. In answers to parliamentary questions as the CAR situation became tense with news of rebels moving south towards the security corridor manned by Chadian soldiers filtered through, government indicated further that there were two components of the agreement: general capacity enhancement and VIP element.

What government could fairly be criticized for is the logic of the very decision to renew the agreement and the extent to which this was state-to-state cooperation or be reduced into a personal protection for a beleaguered president. The decision to renew to VIP component was even more problematic. Government could also have anticipated that Bozize who had already made a desperate call to “brothers” in the west for protection would use our presence on the ground to survive the rebellion merely as a form of regime security.

At the operational level, knowing all the time about that Seleka were stronger than government offices having almost deposed government in December, South Africa could put strong intelligence capacity on the ground to anticipate the growing anti-South Africa feeling among rebels and their decision to attack the South African barracks in Bangui. Thus, the death of 13 soldiers who fought gallantly against about 3000 armed rebels could have perhaps been avoided.

Then, after the tragedy of Bangui, the government simply lost its plot to a large extent due to failure to understand the South African society. It failed to understand that with elections a year away any misstep, even if really innocent, would become useful political football. It seems not to have understood the need for coherent communication by experts and not various voices representing government, ranging from ministers (who should never play the role of spokespersons because they are not communication specialists) to spokespersons of the Ministry of Defence and of the SA Defence Force. Defensive communication is not an optimal way of dealing with a crisis such as we had. An opportunity to be fully transparent to the point of quoting directly from agreements and releasing the defence force’s own pictorial and audio-visual material to put facts across was also not used.

There was failure to engage pro-actively with civil society. There could have been independent analysts going to the ground to understand matters better. In the end, therefore, you had a government consigned to a corner by its critics, but also by its own approach.

The decision not to pull soldiers and equipment quickly out due to the shock of deaths and public criticism was wise as it sent the signal that South Africa had enlightened interests in the region. It had to ensure that it was not seen to be deserting the region and needed to consult regional powers now that there was no legitimate government in Bangui. It was wise to announce the withdrawal at the ECCAS Summit and to do so not out of narrow national interests (to limit further loss of soldiers), but because there was no legitimate government to do business with as the rebel take over had been rejected by the AU and the ECCAS.

Now, the normalization of relations announced over the week of 27 April was explained in terms of ECCAS legitimation of a transitional government under Tiangaye as had been ordered by ECCAS Summit a few weeks ago. But this is still dicy as the Bangui government remains a product of a military take-over. What does this do to the AU principle against unconstitutional change of power. If the aim is to relate with the rebel government in order to nudge it to keep its promise to return government to civilian rule, then this must be made plain before South Africa is accused of lacking diplomatic spine and commitment to constitutionalism in Africa.