The Perspective
Atlanta, Georgia

October 26, 2004

Introduction
In part I of this paper I discussed the issue of pre-election governance reform. Some Liberians are calling for the postponement of the scheduled 2005 elections to enable Liberians institute reforms to correct "flawed" governance institutions. It is argued that we will be reinstating these institutions if elections were held without instituting the necessary reforms. The advocates for pre-election governance reform are calling for the holding of a national conference to agree on the reform measures.

I argued in part I of this paper that the foundational problem that needs to be solved in Liberia has to do more significantly with people rather than institutions. In my opinion, we will not make significant progress in solving our problem of governance as long as those who man our institutions do not conform to the laws of the land, and those who are charged with the responsibility of ensuring that they do either don’t enforce the law, or are themselves unprincipled.

I argued further that even if a national conference were deemed necessary, current conditions on the ground in Liberia are not conducive for the holding of such a conference at this time: there is no demonstrable will on the part of those who would be counted on to implement the results of the conference to do so; those Liberians who are now returning home will need sufficient time, no less than a few years under the best of conditions, to settle down; and many Liberians are not expected to return home during the next year or two. During this period of resettlement, I thought it would be prudent to concentrate our resources and energies on minimizing the burden of rebuilding. Holding the elections in 2005 as scheduled, I argued, would enable the country to move forward with a government that is at least accountable to the people, and thus pave the way for most Liberians to return home.

In this paper I discuss the issue of census. The need to conduct a population census is one of the other major reasons advanced for proposing the postponement of the scheduled 2005 elections. I first discuss the arguments in favor of postponing the elections in order to conduct a population census and then discuss why I think it would be prudent to delay the census, not the elections.

Case for Conducting Census Before Elections
The most compelling case in favor of postponing the elections in order to conduct a population census has been made by Counselor Tiawan S. Gongloe. Counselor Gongloe has spoken and written extensively on the subject but I think his article, "Elections in Liberia without Census Would Be Unconstitutional," published by The Perspective web magazine on May 20, 2004, is perhaps his most detailed treatment of the subject. For the sake of discussion, I make extensive reference to the ideas espoused in this article, which I think summarize the general thinking on this subject.

Counselor Gongloe makes two major arguments in support of postponing the 2005 elections in order to conduct a population census. The first, and perhaps his main argument, is that census must be conducted to ensure that the elections are held in line with the constitution. The second argument is that conducting a census now will afford the country the opportunity to acquire essential data for proper planning and national development. I summarize below the salient points of his arguments as expressed in the article referenced above:

Census Required to Determine Eligible Voters.
Counselor Gongloe quotes article [80](c) to set one of the major constitutional requirements for a public election in Liberia. That article reads in part: "Every Liberian citizen shall have the right to be registered in a constituency, and to vote in public elections only in the constituency where registered..." In order words, registration of voters cannot be done without first determining constituency boundaries----clusters of approximately 20,000 people----since each voter can only register within a known constituency.

Having set the constitutional requirement, Counselor Gongloe then argues that there are currently no constituencies in Liberia and, therefore, it would be unconstitutional to hold elections without first demarcating them. This, he argued, would only be done by conducting a census. He wrote: "Are there known constituencies today in Liberia? ...Is there any Liberian today, given the massive population movements due to the Liberian civil war, who can identify a constituency anywhere in Liberia? If not, then, given the definition of constituency enshrined in the Liberian Constitution, can constituencies be demarcated by NEC (National Elections Commission) without the holding of a census? Note that the constitution clearly linked the definition of a constituency to the availability of census data."

Census Required to Equitably Determine Representatives to the Legislature
Unlike the number of senators who are allocated two per county, the number of representatives to the national legislature from each county is a function of the population (number of constituencies) of the county. Each constituency is entitled to one representative.

Counselor Gongloe argues that because of population growth since the last census in 1984, and because of massive movements of people as a result of the civil war, it would be impossible to determine the number of representatives from each county without a census. A census, he argues, will also determine the number of representatives that each of the two new counties, created after the last census, would be entitled to. And since the two new counties were carved out of existing counties, the counties so affected could have their number of representatives reduced. Counselor Gongloe wrote: "Liberian population has experienced massive movements. Also ...new counties have been created. The constituency boundaries within these new counties have to be determined by the NEC using a new census data based on population growth and movements. Further, the creation of new counties might lead to a reduction of the number of constituencies in some counties." Counselor Gongloe is concerned that we run the risk of over or under representation if elections are held without the accurate population data that a new census would provide.

Census Required for Planning and Development Purposes:
Conducting census before the next election will afford the country the opportunity to acquire valuable data for planning reconstruction and development programs. Counselor Gongloe wrote: "First, the enormous support that is being provided by the international community for peace in Liberia offers the best opportunity to do a proper census, not just for the ensuing elections, but also for obtaining essential data for a proper planning of national reconstruction and development."

Why Census Should not be an Either-Or Issue
The issue of census is one that is difficult to argue against. Census is a clear requirement of law and our system of representative government rests heavily on the size and spatial distribution of the population. The representation allocated to a political subdivision in the House of Representatives is a function not of the landmass that that political subdivision covers but of the number of persons who inhabit that land mass. As previously noted, representation in the senate is allocated equally amongst the counties, irrespectively of population size.

Based on Counselor Gongloe's major arguments, it is easy to conclude that the issue of census is purely a legal one: What does the constitution say? How much of the constitution is currently in force? What is the status of the National Transitional Legislative Assembly (NTLA) vis á vis an elected Legislature, since it is the Legislature that is empowered to cause a census to be held as stipulated in Article 39 of the constitution? The article reads: "The Legislature shall cause a census of the Republic to be undertaken every ten years." But I think there is more to the debate about a census than just satisfying some legal requirements. Dr. Sawyer rightly observed in his September 7, 2004 article titled, "The Need For Pre?Elections Governance Reforms: Continuing the Dialogue" and published on The Perspective web magazine, that our system of governance has collapsed. That means, I think, that our current system is not entirely in line with the constitution. It seems to me, therefore, that the issue we are grappling with, and which is really what this debate is about, is how to reconstitute the system, employing means and methods that are as close to the constitution as practically possible.

Counselor Gongloe's May 20 article was prompted by a May 7, 2004 article carried by The Perspective web magazine. In that article, the Chairman of the Elections Commission was quoted to have said that conditions in Liberia were not right for conducting a census and, therefore, no census will be held before the 2005 elections. The Commission reportedly reached that decision based on an assessment conducted by the United Nations team working in collaboration with the Commission and the Ministry of Planning.

That current conditions in Liberia are not right for conducting a census cannot be debated: Many Liberians are still displaced, both externally and internally; the UN is still settling Liberian refugees in these United States; and many towns and villages are yet to be rebuilt. I think it would be difficult, if not impossible, to get any meaningful data for planning and development, or draw up realistic constituency maps under these conditions.

Assuming that we all agree that conditions are currently not right to conduct a census, we then have two basic options, vis á vis the next elections. One option is to delay elections in order to improve conditions, conduct the census after it has been determined that conditions are satisfactory, demarcate constituencies, and then hold elections. I think this was what Counselor Gongloe had in mind when he wrote: "...the full and proper coverage of each item on the peace agenda, not time, [emphasis is mine] should be given the highest consideration in the implementation of the CPA."

But what would be a reasonable, and that is the key word, reasonable, time period within which all of this can be done, one year? Can it be reasonably expected that conditions in Liberia can be made right within a year for conducting a census, or do we need two years, or even three years? We must keep in mind that the whole peace process is dependent on other people's resources.

It seems to me it would take more than a year or two to create conditions that would render any census meaningful, assuming there is the commitment on the part of the political leadership. Resettlement would involve more than just being physically present in one's home region. I think it would also involve rebuilding towns and villages, and at least being able to once again feed one's family without international relief assistance.

The second option, assuming we all agree that current conditions in Liberia are not right for conducting a census, is to determine a sensible, less-than-ideal means of answering those questions that a census is meant to answer, as it relates to the holding of national elections. The major question, as I understand it, has to do with determining the number of representatives to be allotted to each county, including the two new counties. There has to be a reasonable way of making this determination short of a head count.

One suggestion that is already being discussed is to use the method of proportional representation, as was done for the 1997 elections. While Counselor Gongloe does not support this method, he admits that it is a good system that has been successful in other places. He wrote: "Proportional representation has worked in many places; therefore, the issue is not whether or not it is a good system." One of his concerns with the system is that its first use in Liberia produced "bad results". He concludes, therefore, that "...many Liberians don't trust it (proportional representation) anymore or at least, are suspicious of it."

If it is agreed that the system of proportional representation can produce good results, perhaps what is needed is a determination of why its use in 1997 yielded bad results. We may be able to correct the problems so that its use this time will produce the desired results.

But proportional representation would not be my personal preference, partly because I am not sure I fully understand how a "good" proportional representation system would work in the Liberian context. It seems to me we will have little or no say in selecting our representatives.

Another idea to be considered for determining the number of representatives to be allotted to each county is to conduct the elections based on the constituencies that existed prior to the war. This would be my personal preference because I think this method would bring us much closer to the constitution. The worse that can happen if we employ this method would be that some political subdivisions may be over represented. With so many Liberians killed during the war and several still out of the country, I would be surprised if any region would be under represented based on the old constituency map. And if we must err, it seems to me it would be better to err on the side of over representation than under representation.

There are groups in Liberia who already assume that the next elections will be held based on the old constituency demarcations. The Analyst web magazine reported, in an October 18, 2004 story, for example, that Mr. Edwin Snowe, Jr. was petitioned by the people of constituency #4 in Montserrado County to represent them in the House. Without new census data, these people assume that that constituency still exists.

The one difficulty with using the constituency maps that existed before the war is how to treat the two new counties that were created after the maps was developed, and the counties from which they were carved. How does one determine their constituencies, and therefore the number of representatives to be allotted to them?

One of the new counties ----River Gee----was carved out of Grand Gedeh County, and the other ---Gbarpolu-----was carved out of Lofa County. Before attaining county status, these areas had a known number of constituencies. In the case of River Gee, formerly known as Lower Grand Gedeh, for example, there were two constituencies. Today, it is quite possible that some citizens of that region who resided in the then Upper Grand Gedeh, now Grand Gedeh County, have returned to the region because of the war, and also because the region has now attained county status. But I am not sure if such movement would significantly change the constituency map of the region if it was accurately drawn in the first place. Keep in mind that a constituency is a cluster of approximately 20,000 people. During this interim period, it would seem logical to assume that River Gee would be allocated two constituencies and the number of constituencies within Grand Gedeh County reduced by two. A similar accounting could be done for Gbarpolu and Lofa Counties.

While using the 1984 constituency maps is obviously not the ideal, it would at least enable Liberians to have a say as to who represents them, and thus enable the country to move forward. Census must not be an either-or issue - either we conduct a census or no elections. We must work within the existing legal framework and governing institutions to take measures that would enable the country to move forward, short of a head count.

Conclusion
The issue of a census, especially as it relates to verifying and demarcating constituencies, presents a real problem for the scheduled elections, but I don’t believe this is a problem that reasonable people cannot resolve. Where our country is at today, census must not be an either-or issue. We must find a reasonable means of obtaining the information that a census would provide relevant to the elections. I think many Liberians believe that electing a government that is accountable to them is crucial to beginning the real work that needs to be done in the country---reconstruction, reconciliation, search for justice, reform, etc.

Considering all that is happening in the country today, it is my opinion that conditions in the country cannot be made significantly better in a year than they are today. Therefore, delaying the elections for a year is not likely to bring us any closer to conducting a meaningful census. In my opinion, census should not be held sooner than four to five years following the 2005 elections. By then, things would have settled down sufficiently enough to conduct a census that would provide useful and reliable data. I think viewing the 2005 elections as the beginning of the real transitional period may help us not to insist on creating ideal conditions before the elections.

Concerns have been expressed about the possibility of a future conflict arising from an election that was held without accurate census data. Counselor Gongloe wrote: "... let the current peace process proceed in a manner that would exclude any rational basis of an excuse for another set of criminal gangs to terrorize the Liberian people for political power." This and other similar wise counsels arise out of experience gained from the 1997 election. I think the caution and concerns are well founded. However, I think some of the conclusions being reached from this experience do not seem to take into account the totality of the Liberian conflict.

Charles Taylor cannot be viewed as your ordinary corrupt dictator. In my opinion, Mr. Taylor was a leader who seemed to have believed that all that mattered was power and control, and that no means was too inhumane or unacceptable in the acquisition and maintenance of that power. There may be many more such men in the country, but it does not seem to me that any of the individuals seeking the presidency now has the means, as Taylor did, to run anyone out of Liberia after being elected president. Moreover, none seems to have the grand desire of destabilizing the whole West African sub-region in order to control it, as Taylor did. It is important to consider that there would have probably been no ULIMO had Taylor not gone into Sierra Leone, or LURD had he not gone into Guinea, or even MODEL had he not gone into the Ivory Coast.

While the possibility of some criminal gang springing up cannot be overlooked, it might be safe to conclude that the chances of that happening primarily because elections were held without updated constituency maps may be somewhat remote, especially if the elections are free, transparent and fair. The real challenge and a potential source of tension, as I see it, is the possibility of non-Liberians, especially some of our neighbors from Guinea, Sierra Leone and the Ivory Coast, voting in the next election. This is why I think registration of voters would be crucial, and I am sure that the NEC can use all the help that it can get to deal with this and other potential problems.

---