Chaiwat Satha-Anand is professor of political science, Thammasat University, and chairperson of the Strategic Nonviolence Commission, Thailand Research Fund has an op-ed in the Bangkok Post stating “Census holds key to defusing stalemate”. He states:

Elections, admittedly with all their imperfections, are one of the most significant innovations in conflict resolution because “we the people” throughout history have found a way to solve the thorny political decision of who should rule by counting “our” choices through the voting system. Works on democracy focus primarily on the question of “who” should be counted.

But an often neglected principle of this development is the fact that the process of counting has indeed become a technique to solve this conflict of governance, to determine where sovereign power should reside, at least for a period of time.

The success of democracy as a conflict resolution process has been to transform numbers of arms and manpower into numbers of ballots and voters. But counting remains a decisive measure of who should win in such a conflict without spilling blood or killing anyone, cases of electoral violence notwithstanding.

BP: Nothing here that BP would disagree with, but he continues:

Heads need to be counted, is it possible to use a census as a way to mitigate this protracted conflict, which could turn more violent with the confrontational style of primarily non-violent street protests?

Taking a census is not the same as a referendum, which cannot be undertaken at the moment since it is covered by Section 165 of the constitution and would thereby be problematic for the precariously placed Yingluck government. It is not polling people’s opinions, where both polling methodology and pollsters’ credibility have already been seriously called into question.

Since the 19th century, a real innovation of census taking has been in its systematic quantification. It is not asking for a person’s opinion but their information; much of the time, his identity in terms of gender, age, religious affiliations, income brackets, among other things.

I would argue that, with proper design and relying on bureaucratic professionalism, the National Statistical Office has the expertise to carry out such a census within a short period of time, and that even though the cost of a nationwide census covering more than 20 million households would not be cheap, it is far lower than the cost of an election or referendum.

If the National Statistical Office is ordered to organise a census on the question of “Which side are you supporting — the PDRC or the UDD?”, the answers collected will result in a process of counting the number of people behind each rival group, not unlike asking for people’s religious affiliation or gender identification.

There could certainly be several choices for respondents to choose including “supporting both” or “supporting neither” or “not recognising both sides”. Census takers could visit people’s residences to hear their opinions without the need for protesters to take to the streets.

These answers would be counted as a credible outcome of the numbers game desired by both conflicting parties without the risk of violent confrontation. We might also learn how divided society has become and how big or small the group is which remains neutral.

Once the number of supporters are counted, both movements and the forces behind them could properly assess the outcome of this protracted conflict, not unlike counting the number of people who come out onto the streets, but more comprehensive and more credible.

BP: With the greatest respect to Ajarn Chaiwat, who BP normally finds interesting to listen to or to read what he says, he doesn’t appear to have really thought through his proposal (or at least not explained it). BP is unsure if he is suggesting this in lieu of an election or just in addition to.

First, there is the timing issue. A census normally takes some period of time to prepare for. Actually it is usually years, but then they know this in advance so are planning for a long time. One imagines the NSO could shorten this process considerably based on the knowledge they obtained from the 2010 Census, but it is going to take them a few months, at least, to prepare (they will need to request a budget, the government and then the EC will need to approve etc).

A census takes a period of time to collect the data. From the NSO Web site for the 2010 census:

The 2010 round of the Census is being implemented during September 1 -30, 2010.

It also take a period of time to collate the data. From the NSO Web site for the 2010 census:

Preliminary Report: 3 months after completion of field work
Advance Report: 6 months after completion of field work
Final Report: During 2011 (province-by-province reports)

BP: Now, if only one or a few questions are asked, it could obviously be quicker for collecting the data and the time it takes to collate the data, but you are looking at least 2-3 months and this is not including preparation time. Remember it takes time to visit people’s houses (the 2010 Census included telephone and the Internet as well). BP would be surprised if would take less than 6 months, and probably longer.

Second, the sanctity and reliability of the information. There is little benefit in trying to manipulate census data so no need to seal the information, but if we are going to make political decisions based on the information and we have enumerators going to people’s houses to collect the data, won’t this be open to manipulation? Also, will people trust enumerators? It is not a secret ballot. It is one thing to tell people age, gender, and other details, but if they start asking a political question on who you support, will people feel comfortable telling the truth? Also, how can we be sure that the enumerators will list the information correctly or won’t change it? If the actual result was to matter then there would be a huge incentive to manipulate the result. There is a reason why there is a secret ballot for elections and referendums and that is ensure the sanctity of people’s decisions.

Third, once we know how many people are on either side, what will it mean? Let’s say 25% support the reds, 20% support the PDRC, and 55% support neither (or reverse the reds and PDRC if you want). Then what? Should the reds or the PDRC then have a greater say if they have more than the other? There will be many who don’t support either, who will represent them. Simply knowing how many are in each camp doesn’t mean much unless there are consequences. If it is merely to count how many are in each camp and it does not matter, BP can understand the rationale for wanting to do this, but am unsure how it really then solves or resolves anything. The problems in “second” above about sanctity of information means it will be disputed. Just imagine the “census buying” allegations…

btw, did wonder about the cost, for the 2010 census it was 900 million baht, but BP wonders if this would cost significantly more to better try to protect the data.

In BP’s view an election is a better way to find who supports what. Parties can propose policies and what they will do regarding constitutional and electoral reform. If we really need to know who supports the PDRC, UDD, or neither side, then a referendum* once we have a new government would seem a better and quicker way.

*If the wording can be done legally