There is perhaps no political term that is tossed about more liberally than “landslide”, and applied to political contests that cannot satisfy that definition.
This definitional confusion arises largely from two sources, one of which is the fact that few political scientists in this region have expended much energy articulating a clear and precise definition of the electoral landslide.
The second and perhaps more legitimate reason relates to the fact that our islands are politically peculiar since we are smaller with fewer representatives. As such, a shift in the voting behaviour that would be nothing more than a ripple in the United States or Britain can bring about change of seismic proportions in the Caribbean. The gravity of our situation is perhaps best illustrated by reference to the case of St Kitts and Nevis with a Federal Parliament comprising ten seats; hence a shift in any three could easily be classified as a landslide.
Well established political dictionaries and encyclopaedias such as The Oxford Dictionary Of Political Terms and Blackwell’s Encyclopaedia Of Political Science curiously omit this term, but liberally use it in reference to other political issues and terms that are defined. Reference to normally less reliable Internet sources reveals one rather useful definition which may be found at http://www.wisegeek.com.
This defines a landslide as “an election in which one candidate wins by a substantial margin. If a candidate achieves a landslide victory, it suggests a strong mandate from the people”. This is perhaps a good starting point for discussion since reference is made here to a “candidate” while our references are more commonly to a “political party’s” impact on the electorate.
The definition’s reference to “substantial” also speaks to the inherent challenges associated with defining this concept which is clearly subjective and therefore resists a precise definition.
If, however, there are those among us who desire greater precision, one might want to consider a Wickham(esque) definition that speaks to an election which:
• produces an abnormally high electoral swing; and
• results in an abnormally high seat gain or loss for one political party; and
• results in a fundamental change in the political landscape (not just a change of government).
Any election which satisfies at least two of these three conditions would, in my opinion, be a landslide.
Against this background we can therefore explore the case study of elections in Barbados and exploit the appended table which conveniently identifies the critical statistics since independence.
It is immediately noticeable that this table pays no attention to constituency contests, largely because no single contest could be described as a “landslide”.
Even if one were to allude to the February 1976 St Philip North contest that was perhaps our most dramatic constituency battle, it could not be argued that this seat gain by the Barbados Labour Party fundamentally changed the political landscape, although it did forecast approaching “doom” for the Democratic Labour Party.
It is also noticeable that historically eight seats have changed hands on average in our general elections and in most instances the majority of these have not favoured the Government.
In addition, the swing has been on average – nine per cent away from the Government. It is therefore easy to classify the general elections in 1971 and 1999 as the best examples of electoral landslides since they satisfy the conditions identified.
In both instances the direction of the swing and, indeed, its enormity were dramatic. In the case of 1971 the additional seat haul is identified as “1” which is to some extent misleading since the DLP did gain several seats which did not previously exist and as such could not be said to have “changed”.
In the case of the 1999 election that one fundamentally changed the political landscape since it pushed the BLP’s average support since independence ahead of the DLP’s for the first time and this “perch” will make it considerably easier for the BLP to win elections here for some time to come.
The other three elections that can be classified as landslides would be the 1976, 1986 and 1994 elections and it is not surprising that in all three instances governments were changed. This could give rise to the assumption that whenever Barbadians change governments there is a landslide, were it not for the 2008 election which brought change, but with a swing that was too moderate to be classified as “abnormally high”.
Similarly, there was no “fundamental change” in our political landscape in 2008 since the DLP has not now captured any traditional BLP ground that was previously denied it (such as was the case in the 1976 by-election).
Based on this analysis therefore, there have been five genuine electoral landslides and the BLP can take credit for three while the DLP can take credit for two. Both the DLP and BLP can boast of scoring landslides while in office (1971 and 1999), while the DLP is distinguished as the only party which gained office in an election where the victory was not a landslide (2008).
Although the outcome of the next election is not yet known, the trajectory based on the May 2012 CADRES poll implies that the 2013 election will also not be a landslide and as such the BLP would therefore also join the DLP as a party that gained office without a landslide in its favour.
Needless to say, contemporary political developments could fundamentally change this assessment as either the BLP or DLP could dramatically improve their performance.
• Peter W. Wickham is a political consultant and a director of Caribbean Development Research Services (CADRES).