EARLIER this week, the SNP caught Labour in an opinion poll for the first time. The

System Three survey for The Herald put both parties level at 40%. This was on voting intentions for a Scottish Parliament. As Donald Dewar, the Scottish Secretary, was quick to point out, Labour had a substantial 20% lead over the SNP on voting intentions for Westminster. This throws up an anomaly which needs further exploration. How can it be that support for the party which advocates independence fluctuates from 28% for Westminster elections to 40% for Scottish Parliamentary elections?

When the composition of the support is examined more closely, then some patterns begin to emerge. One of the most important is age. In the 18-35 group, 37% support the SNP for Westminster; however, this increases when intentions for a Scottish Parliament

are considered.

In the 18-24 group, SNP support is at 39%, but among the 25-34 age group support for the SNP leaps to 46%. Younger people are therefore turning to the SNP in greater numbers for a Scottish Parliament than for Westminster.

There may be a number of reasons for this. It is certainly a trend that has been seen before. When the SNP support first surged in the late 1960s and early 1970s, it was in parallel to a wider trend of weakening party loyalties. A new, more socially mobile, more affluent and more educated younger section of the population had less need to rely on party loyalties passed down from generation to generation. This phenomenon, known as partisan dealignment, resulted in a softening of traditional Labour and Conservative party support, particularly among younger voters.

With a further expansion of higher education over the past 10 years, with voters who came on to the electoral role during the latter half of the Conservatives' period of office from 1979 to 1997, then it is perhaps not so surprising that both Labour and the Conservatives are vulnerable. The SNP is not tainted with governmental failure or sleaze or many other of the problems which have beset the major parties in recent years.

The weakening of party attachments has led to an increase in what some analysts have termed ''issue voting''. Rather than voters expressing loyalty to a party through the ballot box, issues and policies are considered, and voting becomes a more instrumental process.

Any voter with a policy goal

of independence for Scotland would make a vote for the SNP possible even if other SNP policies were not agreed with. It is therefore not necessarily contradictory for voters whose natural sympathies may lie with other parties to consider voting for the SNP. After all, in an independent Scotland it would clearly not be the case that the SNP would be the only party on offer to voters. Scottish equivalents of the other parties would almost certainly emerge after which voters could return to a party with a policy profile more closely matched to their own preference. This could be a possible explanation, in part, for the apparent contradictory nature of a large number of Conservative preferences (23%) shifting from the Tories to the SNP when answering the Scottish Parliamentary question.

A further line of investigation must be the idea of voting context. That is the perceived purpose of the election. For a General Election the context is well established for voters, whereas for the Scottish Parliament it is yet to be developed. It is clear that voters are to a degree distinguishing the two elections. On that basis, it is not contradictory to support one party in one context and another in a different one.

We see this phenomenon in other elections, particularly in by-elections, where often a protest vote is more acceptable because the context of the election is not about choosing the government of the day. In the 1989 European elections, 15% of voters voted

for the Green Party at a time when environmental issues were high up in the political agenda.

Before and since, the Green Party has failed to make that sort of impact in any other national election. In a Scottish Parliamentary election, the future of Scotland is a more visible aspect of the context than it might be in a General Election. For many years now, writers have noted the somewhat schizophrenic nature of Scottish politics, or the ''Caledonian antisyzygy'', as Tom Nairn once put it. Once more it is shown to be the case. For a number of years now academic monitoring of Scottish attitudes to whether they feel more Scottish or British has shown this.

In 1992, in a survey of the Rowntree Foundation, 37% of Scots felt Scottish, not British, while 27% felt more Scottish than British, and a further 25% felt

equally Scottish and British.

The idea of a dual identity has fostered dualism in politics as well, and in certain circumstances voters express British choices and in others they express Scottish choices.

For Labour the challenge is to keep the two in line with one another. However, it may be having to contemplate the idea of the end of British politics.

n Malcolm Dickson is a lecturer in politics at Strathclyde University.