The linguistic varieties of the UK and Ireland are discussed along geographical lines. This arrangement by region is convenient in terms of structure, and is helpful to the user who wishes to understand regional differences, or who needs to concentrate on the variety or group of varieties found in one particular region. But it is also potentially misleading, since the impression might be gained that UK and Irish varieties are tidily to be separated from each other, with one being spoken by a fixed, geographically identifiable group of people quite distinct from another group using another quite different set of speech-forms.

Nothing could be further from the truth. Far from there being regional cut-off points for ways of speaking, i.e. boundaries where, for example, one accent ceases to be heard and another takes its place, accents and dialects blend subtly and imperceptibly into one another. Rather than the hearer detecting the presence or absence of features as they move about a country or region, particularly at a local level it is a matter of ‘more or less’, of features being heard with greater or lesser frequency as features most characteristic of one region are left behind, to be replaced with greater intensity by others associated with a region being approached.

Nor should we think that all speakers in one place use the same set of features with the same level of intensity, if they use them at all. It is to be expected that some speakers, those who sound most local to a particular place, will fairly consistently exhibit a set of features which most closely conform to a characteristic local way of speaking, and it is these which form a central part of the local accent and dialect descriptions given later. However, very many speakers will not be consistent in their use of these features, being variably more or less regional in different situations or under different social promptings (e.g. the social status of addresser and addressee, and the degree of familiarity between them), even within the same discourse (e.g. depending on the topic). It is important to note immediately that such variation is not random: speakers do not drift between, towards, or away from markedly regional pronunciations on a whim. Rather, it has been shown in numerous studies that such movement patterns correlate with such social phenomena as age, gender, socio-economic status, ethnicity and local affiliations of both speaker and hearer, and can result in short-term, but also long-term, language change.

The acceptance of the absence of tight boundaries for phonological and grammatical features, and the acknowledgement of speakers in any one place being socially heterogeneous and, moreover, inconsistent in their speech lead to the inevitable conclusion that the concept of the ‘dialect area’ as a fixed, tidy entity is ultimately a myth. In terms of pronunciation, what we are faced with, in place of a certain number of accents, is in reality a continuum: accents shade one into another as individual speakers espouse features drawn from a range of accents to which they have access and that are indicative not just of their regional connections but also of their social needs and aspirations. The same is true for grammatical usage, and for lexical choice.