Election 2008 – What Role did the Alternative Media Play? (My Take)

Yesterday’s news yes, but I am desperately trying to avoid work so I will indulge myself. Here goes:

Early morning 9 March, I joked that the election bookies’ books just went subprime, hit by a black swan. No one quite honestly expected the results. Ex-post, many pundits put weights on many reasons as to why it happened. One of them is the role of the alternative media (AM); bloggers, Malaysiakini, etc. The usual explanation is that it provided unbiased information to the people and empowered them with knowledge to cast their votes decisively. To some extent, it is true I guess. But here, I offer a different explanation – AM acted as a new channel that helped to coordinate people’s beliefs, eventually leading to a leap in equilibrium.

Setup

Will throw in a simplifying model to help the exposition. Firstly, there are i potential voters, with their “willingness to vote” denoted by V_i. Here is the important bit, V_i is taken to be:

V_i = f( q_{i ,} E[Q_-j | I] )

where,

• q_i is voter i’s individual drive to vote,
• Q_-j is sum of q_j, j not i
  
• I is information available to i

Basically, this means that any single voter’s “willingness to vote” is dependent on two factors: his innate drive, q_i, to vote and his belief of others' drive to vote conditional to the information that he has, E[Q_-j | I]. It is taken that the innate drive is fixed for an individual but varies across individuals – that is you have some people with high q_i , like say, Nat Tan, and many people with low’ish level of q_i , perhaps like your apathetic cubicle-mate. Then, there is E[Q_-j | I] which basically says that a voter’s willingness to vote depends on others’ willingness as well.

A rough example would clarify things further. Consider the KL Freeze that happened not too long ago. Your decision to freeze in the middle of Pavilion would be dependent on your wackiness level, and if you believe others will be freezing too. For example, if you belief no one will freeze, unless you are superbly wacky, you won’t really be too keen to freeze alone in public. If you believe 5 people will be freezing too, then you might be tempted to join in if you are sufficiently wacky, or not if you are not. If you believe 100 people will be freezing, chances are you will be freezing too even if you are wacky at a normal level.

Here is the cool part – now consider the collective level. If you believe that 100 other people will be freezing, you will freeze too. At the same time, every individual from the 100 people believe that the other 99 plus you will be freezing, then in total, 101 will be freezing. Self-fulfilling prophecy at work. Then consider that there is another individual who is wacky at a slightly less than normal level, say, she will only freeze if she believes 101 people will freeze. Now, she will freeze too, since you decided to freeze. Then consider another person who is even less wacky who will only freeze if 102 people do the same. Given the person before will be doing it, and he believes she will do so, then he would freeze as well. And so on, so forth as the cascading effect continues.

This explains how a group of very different individuals could end up all freezing as long as there are sufficient people who are wacky enough. Call a tipping point, leap in equilibrium or critical mass cascade if you want. Note however this might not always apply. If the group of people has 1 extremely wacky guy who will always freeze and the rest being much less wackier who will only freeze if more than 20 people are going to freeze. In this case, the wacky could freeze all he wants, and he won’t be able to compel next guy to freeze too, so no cascading effect will happen. Unless, of course, everyone in the group could coordinate before hand, and make sure everyone freezes.

How does this Relate to the Election? (Coordination)

Well, I guess the KL Freeze example is sufficiently clear to suggest how this relates to the election. Assume that there are many potential voters whose q_i is such that they are leaning towards voting the opposition (you can think of all the reasons why). However, in the initial condition, E[Q_-j | I] is such that most believe that the other guy won’t be voting for the opposition, or won’t be voting period. Most expected BN to win even if they put in baked potatoes as candidates (and they did in some cases…) in the initial state. This is due to the I bit of the equation – everyone has a prior belief that nothing will change even if they vote because everyone else won’t be doing the same anyway, so their overall willingness to vote is low, even if their innate willingness is there.

Now this is where AM comes in – they updated people’s E[Q_-j | I] by updating I. At first, some vocal bloggers started to blog about his intention and willingness to vote, providing a signal to the rest of people that Q_-j could be higher than expected. And using the same analogy as the KL Freeze above, this led to the next slightly less vocal person to voice out too and signal to others, and the chain reaction began, creating a ripple effect that eventually turned I, hence E[Q_-j | I] upside down. Starting from mushrooming of individual blogs revealing people’s dissatisfaction and willingness to vote, to Youtube videos of tens of thousands of Bersih/Hindraf protestors on the street getting shot by water cannons, to massive ceramahs by the Pakatan Rakyat such as the Penang ones, the silent majority of the voters were embolden and found themselves not in such lonely company, i.e., E[Q_-j | I] tipped from extremely low to sufficiently high[1]. This led to V_i being tipped as well from a low equilibrium to a high one. And thus the black swan was born.

Note how this put the importance into the quantity of people signaling, i.e., lonely under-read blogs matter too. The big hubs of information, like Malaysiakini and MalaysiaToday and so on are important in conveying quality of information perhaps, but the main coordinating effect emphasised in this model is in the commentators, letters to editor and ‘vox populi’ and the such – the Web 2.0 bits of the hubs, as well as the pictures of thousands of people supporting the cause. It is like, oh my gosh, everyone is thinking the same way as I do! And then of course, this updating of I could spill to non-AM realms too through the usual work, family, mamak interactions and so on.

Note that critical in this explanation is how the majority of the voters were leaning towards being dissatisfied with the status quo and wanted change, but just lacked the coordinating incentive to do so in the beginning. Without this, nothing would have happened. Yes, duh.

I mused about Vaclav Havel’s “power of the powerless” late last year, in which a system with the authority hanging on to power without the support of the people is intrinsically fragile:

“…[H]avel urges his fellow citizens to down-play strictly political activity in favor of a strategy he feels will be more successful: cultivating the sphere of truth within individuals in the hope that as this hidden sphere grows, it will becomes an irresistible force that will change society”

Perhaps AM helped the cultivation of this hidden sphere more than we thought.

Elanor

---

[1] But not sufficiently high in provoking a jump into another low equilibrium due to complacency.