Behind Jamaica's Garrisons
(Part 1)
The following are excerpts of the July 1997 Report of the National Committee on Political Tribalism. The Committee was chaired by Political Ombuds-man the Hon. Justice James Kerr.


THE border wars between garrison communities of different persuasions result in:

(i) the increased difficulty in maintaining law and order;

(ii) an ability to maintain social infrastructure (roads, water, sewage, garbage disposal, electricity, shops, supermarkets, markets), which border or pass through disparate communities;

(iii) a restriction of movement through these areas which affects human rights, transportation, and job attendance and opportunities;

(iv) a restriction of business opportunities to the localised area as customers from other communities are denied access by blocked roads and real or perceived threats of violence.


Garrison communities are created by:

(i) the development of large-scale housing schemes by the State and the location of the houses therein to supporters of the party in power;

(ii) homogenisation by the dominant party activists pushing out the minority from within and guarding against invasion from outside; and

(iii) the expelled setting up a squatter community.

The hard core garrison communities exhibit an element of autonomy, in that they are states within a state. The Jamaican State has no authority or power except in as far as it forces are able to invade in the form of police and military raids. In the core garrison disputes have been settled, matters tried, offenders sentenced and punished, all without reference to the institutions of the Jamaican State.

The Garrison Process

The following extract from Mr. Mark Figueroa's important contribution on the topic speaks eloquently for itself and merits quotation in full:

"Encompassed in this process are all those activities that are associated with the establishment, the strengthening and the extension of the influence of garrisons and their associated psychology. In using the notion of a garrison process, I am linking the presence of the large scale garrisons such as Jungle, Payne Land, Rema, Tivoli and other less known strongholds, to a political culture. This political culture extends well beyond the boundaries of the communities which have come under the tight control of politicians, their thugs, and/or local enforcers.

Those who promote the garrison process seem intend on extending its dominance over larger and large sections of the country yet to date there are few constituencies where the vast majority of the territory has been garrisoned. To focus on constituencies alone is to miss the significance of the garrison phenomenon as a whole. The presence of a relatively small garrison within a constituency or the presence of a garrison in a neighbouring constituency can fundamentally alter the course of an election in a constituency that is relatively garrison free. The ability of the garrisons to do 'outreach work' greatly enhances their significance. My central concern is therefore with the concept of a garrison community not a garrison constituency.

Some worked with a notion of eight garrison constituencies. Based on my analysis of the 1993 election results I have also identified eight unambiguously garrisoned constituencies. In addition there are four other constituencies where I estimate that garrison control is at or just above 50 per cent and three where control is between 25 per cent and 35 per cent. In most cases the garrison process is a bipartisan one. That is, in the most garrisoned constituencies vast areas are under garrison control but not by just one party, rather each party has its garrison areas with a number of contested areas in between. In many more constituencies the garrison communities constitute pockets of tight control surrounded by much broader zones that represent contested areas. Here we see the relevance of the concept of a process in two ways. First, there is an effort over time to expand these pockets to the point where the constituency is transformed into a garrison and second, the presence of a garrison within a constituency fosters certain types of actions. In terms of this paper these activities are mainly seen within the context of vote manipulation and efforts to rig the results.

I argue below that the garrison phenomenon has become central to the practice of electoral manipulation in Jamaica. Although the garrisons may in fact be relatively confined political strongholds their significance extends beyond their borders. This can be seen in terms of the connection between the garrisons and two other related aspects of the Jamaican political system.

The first of these is turf politics. That is, the process by which political parties seek geographical or positional control over given areas as an electoral strategy. Beyond this we have bogus voting or electoral rigging. In this paper I am treating the garrison process as the core. Hence I encompass the other related aspects in the discussion as part of a broad strategy based on the building of garrisons."

Mr. Mark Figueroa therefore emphasised, and we accept, that the creation, development and maintenance of garrisons are neither accidents of history nor geography, but part of a process deliberately fostered for political ends. Equally important is his focus on the concept of a garrison community, rather than a garrison constituency, the point being that the existence in a constituency of a garrison community can have as significant an impact on the electoral process as the garrison constituencies themselves, and an impetus to transform the garrison community into a garrison constituency.

Common Features of Garrison Communities

Dr. Barry Chevannes de-scribed the common features in this way:

"Just as the political tribe of ancient Greece and Rome was under the leadership of a demagogue so in Jamaica the garrison towns were controlled by the Dons. "Entry and exit to and from these communities are controlled by the so-called "top ranking" and gang leaders who have close relationships with the constituency Member of Parliament, get preferential access to contracts and jobs and function as key elements of the local level community political leadership in both parties in these inner city poor areas. These constituencies are made up preponderantly of these "garrison communities" where organised political gangs with high powered M-16 and A.K.47 assault rifles and sub-machine guns control clearly defined political boundaries and territories where political protection insulates them from the reach of the security forces."

We interpret this to mean that on occasions political influence has impeded the Security Forces in the performance of their duties.

"There is a link between garrison forces and the party's political leadership which provides two functions:

(i) it provides the main conduit for accessing and distributing scarce benefits. The top rankings thus become the main brokers between the Members of Parliament and the local communities;

(ii) in a transactional sense, the Member of Parliament is sure of retaining his territorial support, while the rankings are able to acquire wealth and local power as well as protection from the forces of law and order" ­ Dr. Chevannes.

We accept this as a correct description of the relationship between the Members of Parliament in certain garrison communities.

Doctors Sangster and Chevannes and Mr. Figueroa all resort to electoral results as a method of identifying garrison communities. On the basis of a 75 per cent of those voting for the candidate of a particular party, Dr. A.W. Sangster in the submission "Point of View" states that between 1989-93 the garrison communities increased from 3 to 11 ­ an increase of 1 JLP and 7 PNP.

Mr. Figueroa's more detailed method of assessing electoral results on a polling-station by polling-station basis and the extent to which homogeneous voting has taken place, is to be preferred as he points out that consistent homogeneous voting in certain rural areas may be based on a legitimate preference of long standing. Despite the differences in methodology, Dr. Sangster and Mr. Figueroa both agree that the following eight constituencies have dominant garrison communities. They are: Kingston, West; St. Andrew, South; St. Andrew, South West; Kingston, East and Port Royal; St. Andrew, West; St. Andrew, East Central; St. Catherine, Central; St. Catherine, East Central.

The Immediate Effects

Among the more visible physical effects are the abandonment of legally owned houses and business premises followed by the capture of some of these by illegal occupants and/or the destruction of others through vandalism and inappropriate usage. By this process, large areas of some garrison communities remain in a permanently derelict condition, which in turn fosters criminality and violence.

Another attendant consequence of the relocation to other communities ­ Portmore and the like ­ would be the considerable economic costs and the unduly heavy burden in those communities in providing for the new residents suitable accommodation with the necessary infrastructure and amenities, while the communities they have left remain with abandoned and unoccupied buildings.

The Consequential and Continuing Effects

The increasing incidence of violence and the attendant space of murders and serious bodily injuries

It is agreed by all that party politics was not the only or may even not be the dominant cause of the record number of homicides in 1996 ­ which was not an election year. The others include drug trafficking and gang related activities. It is, however, beyond debate that party politics was the cradle for factional conflicts, that the political clashes of the late 1960s particularly in the election period of 1967 ushered in the era of firearm offences against the person and that party politics remain a major cause.

Criminal gangs tend to identify themselves with particular garrison communities and remain affiliated or aligned to their particular party. During the election period they surface as perpetrators of political violence and election malpractices.

Area Branding

Residents in garrison communities often suffer from "area branding" in significant respects, including the following:

(a) retaliation from areas of supposed differing party affiliation,

(b) victimisation by employer's averse to violence prone areas or themselves politically biased,

(c) political and economic victimisation should their party not from the government, an immediate risk of losing projects in progress when their party loses and, not least, the general deterioration of market values in real property within these areas.

 Part 2 »


Gives Us Your Feedback




Part 1
Part 2
Part 3
Part 4
Part 5


Kerr Report - Part 1
Kerr Report - Part 2
Kerr Report - Part 3
Kerr Report - Part 4
Kerr Report - Part 5
Kerr Report - Full Report(Part 1 - Part 5)

(You will need Adobe Acrobat to view this file. If you do not have Acrobat Reader, it can be downloaded using the link below)
Get Acrobat Reader

Copyright 2000 Gleaner Company Ltd. | Disclaimer | Letters to the Editor | Suggestions