A Quick Response to Mubarak Leaving: Why Do Protests Ever Bring Down Governments?

In light of the rumored coming resignation of Egyptian President Hosni Mubarak, I thought I would share an interesting conversation with my colleague Peter Rosendorff yesterday. Peter raised the question of why governments ever fall simply because there are protesters in the street? After all, couldn’t governments simply go about their business of governing and wait for the protesters to go home? More generally, as political scientists do we have a good model of what factors actually determine whether protests succeed in removing governments from office?

So to answer this question, we started with the most extreme and obvious explanation, and then tried to walk our way backwards:

  • If protesters actually “storm the palace” and kills the currently leader(s) of the government, then protest will yield government change (e.g., Romania).
  • If protesters stop short of killing the current leader(ship) but somehow forcibly exile him/her/them, then protest will yield government change.
  • If the continued persistence of protest leads the current leader(ship) to think that either Options 1 or Options 2 will come to fruition, then it is possible the leader(ship) will step down and/or flee, then protest will lead to a change in government

But heading back to current developments in Egypt, none of these factors seem to explain what is going on. Instead, we seem to need to get into other explanations, like:

  • If protest convinces other pivotal actors in society that they would be better off with a different government in place, these actors may take actions in order to bring about a change in government. In this case, we could again say that protest has led to a change in government.

So this proposition would force us to identify two key factors. First, who are the pivotal actors in society that can actually force a change in government? Clearly – as is the case in Egypt – the military is always a possible candidate here. But are there others? Party leaders? Key economic figures? Major civil society players? I guess part of what I’m wondering is is this explanation inevitably a story about the military? Or, put another way, will we look back on events in Egypt and describe what changed on Day 17 simply as the military deciding that Mubarak was now a liability, and little more than that?

The second factor this last proposition points to is what exactly is it that protesters can do that convinces these pivotal actors that a change in government is necessary? Is simply bad PR, e.g., protesters getting beaten or killed making it harder for the country to manage its international relations? Could it be more economic, e.g., shutting down commerce in a capital city for an extended period of time? Or might it actually be something more normatively pleasing, such as demonstrating to these pivotal actors that the regime no longer has the support of the people?

I am very interested in responses to this post. For those familiar with “successful”” protests (which I am loosely defining as leading to a change of government), what made the difference? For political scientists who study protests and regime change, what kind of models do we have that explain exactly why protest gives way to regime change? I know we have a vibrant literature on the question of why people protest in the first place (see here for some citations), but what do we know about why protest actually leads to changes in government?


A very brief second observation on the possible fall of Mubarak: based on what I wrote previously on the links between 1989 and 2011, my expectation is that the fall of Mubarak will energize protest in other Middle Eastern and North African countries. Mubarak is a very big fish in the region, I would guess – and this is of course only a guess – that whatever contagion effect Tunisia had is likely to be exceeded by a possible Egyptian contagion effect.


As always, we welcome guest posts from Egypt experts especially if, as now seems likely, Mubarak does give up power later today.

19 Responses to A Quick Response to Mubarak Leaving: Why Do Protests Ever Bring Down Governments?

  1. Chris Hanretty February 10, 2011 at 12:06 pm #

    Not an Egypt expert, but:

    what about the decreasing present value of future rents extracted? If protesters can demonstrate that they would rather the economy tanked than the current government continued in power for one more day, then maybe that would change the net cost/benefit decision for the ruling clique?

  2. NotFromIgnorance February 10, 2011 at 12:08 pm #

    Didn’t Barbara Geddes write a long article examining authoritarian regimes and why they’d be prone to turnover?

    Let’s not pretend this is the first time political scientists have considered this.

  3. Dmitry Gorenburg February 10, 2011 at 12:36 pm #

    It seems to me that in this case, what is happening is a military coup. Military leaders have decided that the protests will not end as long as Mubarak is in power. There may be some fear that as protests spread and strengthen, they will lose the ability to maintain control of the situation. They also seem to feel that people in Egypt still respect the military and may be satisfied with a military takeover.

    It doesn’t seem to me that Mubarak is going willingly. More likely, he was told by the generals that they no longer support him — in other words, he was told that he could go willingly with some dignity or be forcibly removed.

    My guess is that what will follow is an attempt by the generals to control the transition process in such a way that their power and wealth are preserved. Not that different from many other military coups in the last few decades.

    The critical question is whether the protesters will acquiesce in the military takeover (perhaps under pressure from that part of the population that seeks a ‘return to normality’) or if they will stay on the streets to force the military to pursue an actual transition to democracy. Much will be revealed by who is appointed to whatever transitional governing council is put in place. If representatives of the protest groups are included (as in Tunisia), that will be a very good sign. If it’s all military, or military and holdovers of the old regime, that would be a bad sign.

  4. Bill February 10, 2011 at 2:51 pm #

    I have no specialized knowledge whatsoever, but I think you are on the right track with the extreme and obvious explanation. Protests are a way for the population to test the loyalty of the police and military. Governments fall when the police and military decline to shoot the protesters. Once that happens, then the protesters can storm the palace any time they want.

    The storming does not literally have to occur, however. Once the elite (and/or the supreme leader) figure out that the palace can be stormed, it’s in their interests to figure out what they need to do either to get the protesters to go away or to get the military back on board. Or, if they can’t figure anything out, to board the next plane for Geneva.

    So, the resolution of the protest could involve giving the protesters what they want, bribing the military, the leadership class running away, etc. But, the not shooting the protesters is key. Tiananmen failed because the army shot the protesters. Romania’s protests succeeded because only the secret police were willing to shoot the protesters, and there were not enough of them. Hungary’s revolution in ’56 succeeded because the Hungarian army declined to shoot the protesters and then failed because the Soviet army agreed to do so.

    You say this does not explain the current situation in Egypt. So, are you saying that the Egyptian military has called up Mubarak and said “Hey, Hosni, any time you’re ready, we’ll mow them down like corn with scythes of lead” and Mubarak is just not taking them up on it? How do you know this?

  5. David Pion-Berlin February 11, 2011 at 11:59 am #

    The key to understanding why Mubarak just resigned is with the military. Protesters alone almost never bring down tyrants. They need cooperation from the military, meaning the armed forces decision to either remain neutral by neither suppressing the people or aiding and abetting them, but rather allowing them to persist on their own, or to finally step in and assist the protesters by ousting the President in a coup. We know that the military is the most powerful actor in this scenario for the following reason. At some point, Mubarak must have requested that they suppress the protests which were getting out of hand, disrupting the economy and causing the President embarrassment. The fact that they never fired a shot at the protesters is proof enough they defied Mubarak. And yet, not a single high ranking officer–including the Defense Minister–has been fired. That tells us Mubarak did not believe he had enough power to challenge the will of his senior officers. This time, my bet is that the Supreme Military Council told Mubarak last night he had played his last card and it was time to go. Mubarak had no choice but to agree. I’ve studied military strategy during protests aimed at ousting rulers, in Latin America. If you’d like, take a look at my article in the Journal Comparative Politics, July 2010 that argues the military’s dominant strategy is to remain neutral during these episodes. You will find the scenario we (my co-author Harold Trinkunas and I) laid out has striking resemblance to the Egyptian case.

  6. David Pion-Berlin February 11, 2011 at 3:04 pm #

    This was not a classic military coup. I’ve studied the military for several decades now, and this takeover is different. Coups are usually generated out of either selfish institutional interests (salaries, materials, promotional denials, drive for personal gain) out of national security perils (left wing challengers, guerilla movements, etc) or alliances with middle and privileged classes feeling threatened by classes below them. None of those apply here. But that does not mean the armed forces weren’t looking out for themselves; they were. By taking a largely neutral postion over the last two and half weeks, they preserved their reputations, avoided bloodshed which would have tarnished that reputation, allowed momentum to build from the masses below, thus creating an unstoppable tide, and an untenable position for Mubarak. Thus, this takeover is more akin to a populist uprising facilitated by military power .In this instance, military neutrality obviously paid off for the protesters, though it could have gone differently. Of course, while it is likely the military will not now serve as a caretaker government, paving the way for elections, it is not certain. There are always the temptations of power, and the desire among some generals to stay. Hopefully and probably, they will be overriden by those who see this takeover for what it is: propelled by popular demands for democratic change, with the military serving as handmaidens, not permanent occupiers of power.

  7. Ronnie finn February 11, 2011 at 4:20 pm #

    Agree with David’s post, it is a popular uprising with the people saying enough is enough and the military carefully safeguarding their behinds. It remains to be seen what happens next…whether any real democracy will follow…

  8. Gweeks February 11, 2011 at 4:33 pm #

    I agree with David that at least so far the situation suggests that, as in countries like Bolivia and Ecuador in the past decade or so, a president is forced out because he loses the military’s support, not because the military wants to rule directly (and not solely because there are protests). Interestingly, despite the fact that Mubarak had relied on the military to remain in power so long, it remained an independent institution and therefore did not feel threatened by his forced departure, unlike many other autocratic regimes. There are a lot of interesting parallels to Latin American civilian-military relations, and I wonder how much research has been don e on that subject in Egypt. Latin America parallels should also caution against assuming a democratic outcome.

  9. Harold Trinkunas February 11, 2011 at 5:23 pm #

    We should also keep in mind the role of the draft in Egypt when it comes to decisions to use force. The Egyptian armed forces use large numbers of conscripts. As one of my military historian colleagues reminded me yesterday, conscription was once seen by liberals in the 19th century has a bulwark against conservative militarism. The logic was that a mass army would be reluctant to repress the mass of the people. This is one of the reasons the left in France preferred a conscript army prior to the world wars. One of the reasons that the 1961 coup attempt in France failed was the refusal of conscripts to support the military uprising.

    If I remember correctly, David and I suggested at least four reasons why militaries would avoid intervening in political crises: fear of reputational damage, fear of future accountability, fear of the consequences of international disapproval, and fear of internal factionalization. This last mechanism may be the one that is most significant in the Egyptian case.

    While there are many differences between the French and Egyptian experiences with mass armies, the fact that Egypt has a conscript force may have contributed to concern within the senior leadership over the reliability of military units if ordered to repress demonstrators. Some of the other factors (reputational damage, international consequences) may have contributed to a sense among senior officers that repression was risky. Under these circumstances, sitting tight and negotiating behind the scenes was the least risky strategy they could have pursued.

  10. Peter T February 11, 2011 at 11:24 pm #

    Couple of comments:

    One – “just governing” is keeping the streets orderly. Persistent mass demonstrations are a way of saying “we can stop you governing”. The challenge is almost never confined to demonstrations – party offices are ransacked, official cars burned, the property of high officials attacked and so on.

    Two – Police/army attitudes are critical, but sometimes protest movements can bring significant pressure on elements in the army or police. Police may share their grievances, army officers may be drawn from the same class. In Iran in 1978, many middle-class families told their officer relatives that it was a choice between the family and the Shah (pretty much all chose the family) – poor families gave the same message to their conscript sons.

    So look at where the army comes from, what its attitudes are, and how connected it is to the protesters.

  11. Dmitry Gorenburg February 12, 2011 at 8:24 am #

    To add to Harold’s comment, which seems to me to be right on, international consequences are also very important for the Egyptian military. Many Egyptian officers considered to have a bright future are educated in US war colleges. As they move up, they return for more advanced training. Furthermore, they get $1.5 billion worth of free US-built military equipment every year.

    They recognized that repression of civilians by the military would cut off those benefits, at least for awhile. Furthermore, US military training emphasizes proper civil-military relations. Officers trained in the US I would guess are much less likely to follow orders to shoot unarmed demonstrators (I’m sure there are studies of this, at least internal to the US military, but I haven’t read any).

    So for these reasons I would argue that international consequences are equal in importance to fear of fractionalization for preventing repression in the Egyptian case.

  12. Charles February 12, 2011 at 1:04 pm #

    I played a modest part as a UK diplomat in the fall of Milosevic in 2000.

    Factors for that one included:

    – getting out fast the election results when Milosevic lost, mobilising the population to push en masse for change and not be cheated

    – quiet messages fed in to the top security elite that their best personal interests would be served by not going down with a sinking ship

    – united Western international front calling for change, with extended training and other support extended to opposition leaders to get them ready for taking power – boosting opposition confidence and so their willingness to take personal risks

    Mostly it comes down to the inner people around leaders simply realising that the game’s up and grabbing what they can before tip-toeing for the exit. Plus sometimes the police simply sit back and wait to see who wins in any given top-level power-struggle – this is what happened in the attempted coup v Yeltsin in Moscow in 1993.

    At root it’s a psychological issue methinks – when do top people used to winning suddenly for the first time in their greedy lives feel raw fear?

  13. Embe February 12, 2011 at 5:46 pm #

    An interesting topic. Though the support of the military is crucial, the military is always smaller than the general population so if the majority of the population is against them (and no longer afraid), the protesters become the most fearsome force in society.

    Also, I tend to agree with Charles (above), the top people (soldiers too) experience psychological fear in the face of the sheer number of unregimented yet committed protesters. Plus no one wants to be on the losing team and when the momentum seems to be moving to the protesters, on the ground, the effect is electric. A soldier would have to be drugged to not feel its pull. If you’ve ever lived through a natural disaster, the Group-Mind that emerges is almost palpable and very difficult to resist. I imagine wars and revolutions are even more intense.

  14. Mark Dionne February 12, 2011 at 8:29 pm #

    Somewhere I heard that the key was when Mubarak lost control of the media. Several important editors stopped pushing his propaganda. Once he lost them, his importance to the military was greatly reduced.

  15. Joel February 13, 2011 at 3:06 am #

    I’ve heard that a pivotal factor was an unwillingness on the part of rank-and-file military personnel to kill fellow Egyptians simply for non-violent political action.

    If soldiers resist orders, it’s difficult to keep the state working. Also, if the military finds itself arresting badge-carrying “law enforcement” on false-flag vandalism missions, that can complicate relations among branches of state power.

    Basically, I think the turning point is when rank-and-file functionaries of the government are placed in a position where they don’t feel they can continue in their roles, and only a change of policy on the part of the government would restore them to a condition in which they will follow orders.

  16. darshan February 14, 2011 at 4:51 am #

    There is a different dynamic involved for violent vs non-violent protests. Non-violent protests work when the government & the protesters have somewhat similar moral values.To give a somewhat older eg., Gandhi’s protests in British India relied on the ability of the protesters to shame a government. Hence the need for common moral values. Obviously you couldn’t have shamed Hitler into giving up power but the British Government in India was a different beast.

  17. darshan February 14, 2011 at 4:52 am #

    There is a different dynamic involved for violent vs non-violent protests. Non-violent protests work when the government & the protesters have somewhat similar moral values.To give a somewhat older eg., Gandhi’s protests in British India relied on the ability of the protesters to shame a government. Hence the need for common moral values. Obviously you couldn’t have shamed Hitler into giving up power but the British Government in India was a different beast.

  18. Jay Ulfelder February 16, 2011 at 6:39 am #

    I think this is a crucial but understudied point. In theoretical terms, the question is this: by what mechanism(s) does the power of popular uprisings get converted into political change?

    The challenge for these uprisings is that they are outsiders, so they have no nonviolent, direct way to remove office-holders or change laws. That leaves them with three options:

    1. Direct violence. This is extremely rare. Even in Romania, the street violence was somewhat ephemeral to the regime change that occurred, in that it wasn’t the rioters who got Ceaucescu and took over, it was security-apparatus insiders.

    2. Coup. Uprisings change expected payoffs for insiders and make them insecure, and this can prompt some of them to take action against the incumbent ruler. As you say, this is the most likely pathway, and it depends on the action or inaction of at least portions of the military.

    3. Persuasion. Sometimes but rarely, incumbent rulers decide it’s time to go without being killed or couped. I think the likelihood of this outcome depends a lot on the ruler’s expectations about his and his family’s personal safety and wealth in a post-overthrow world.

    So, bottom line, I think you’re right that a coup is the most likely mechanism in these cases. It’s not the only one possible, but there are structural reasons why the other two are so rare.

  19. Dan L. February 16, 2011 at 6:11 pm #

    Let’s think of the riots as an external influence on the Egyptian state (though that’s not necessarily what they are). If the riots are because of dissatisfaction with Mubarak and if the riots are causing problems for police then the riots necessarily push apart the interests of Mubarak and the interests of the police. To the extent that a Mubarak ouster would make the police chief’s job easier, Mubarak will no longer have the police chief’s support.

    In general, any persistent system composed of many self-interested individuals requires an equilibrium between the conflicts of the interests of different individuals and the power of those individuals to effect those interests.

    I don’t think you can assume that the protests CAUSE regime change. They’re correlated, but that’s easily explained when you point out that whatever conditions triggered the rioting most likely also disrupted the balance of diverse interests within the system. Maybe the rioting is just the canary.

    Sun Tzu and Macchiavelli both wrote books on how to avoid being in Mubarak’s position by the way. I’d start there if I were looking for research material.