Traditionally, malt whiskies have been classified by region: first Highland/Lowland, then Highland/Lowland/Islay/Campbeltown/Speyside, then a proliferation of sub-divisions of Highland and Speyside. The original division – and to an extent the later sub-divisions – discerned differences in the flavour, style and character to be found in the various regions. Such a break-down was seized upon by writers in the 1980s, when malt whisky began to be more widely appreciated, since it was a convenient way of communicating the virtues of single malt, distinguishing it from blended whisky and begging comparison (for the consumer) with fine wine.With greater understanding of the influence of production and maturation upon flavour – to some extent inspired by the demand for single malts – it has become possible to produce malts with similar characteristics almost anywhere in Scotland. But not quite anywhere. The essential distillery character is unchangeable. But you can easily alter the degree of peating of the malt. You can extend your fermentation times; alter your distilling programme. Vary the wood in which you mature your whisky: has it not been said that this can contribute up to 80per cent of the mature product?So where are your regional differences now, when Islay-style whiskies can –theoretically, at least – be produced on Speyside? And with the increasing interest in individual cask bottlings (which emphasise the difference between one cask and the next), the stress on region becomes secondary to that on wood. There might be other systems of classification more helpful to the consumer who asks, ‘if I like this malt, which others will appeal to me?’‘I classify the malts I use in terms of their style and opulence,’ says Richard Paterson, Master Blender for JBB ( Greater Europe) whose best-selling blend is Whyte & Mackay. ‘I use the traditional four-way regional division [Highland, Lowland, Islay, Campbeltown] and I split my Highlands into heavy, medium, light and floral. The division of Highland malts into First, Second and Third class, is bollocks. It’s a beginners guide, and I doubt whether any blenders follow it today. Dalmore, one of my malts, is every bit as good as a Top Class Speyside for blending purposes. ‘The individual casks in which the whiskies have matured – especially ex-sherry casks – will influence where I marshall the malts; the classification is not rigorous or necessarily consistent. For me, the important things are how the constituents work together, and how they combine after a period of time, the marrying period.‘Selection and classification all depend upon what style of blend you want to create. It is a very personal overview, based on an understanding of the contribution likely to be made by individual whiskies to the overall effect.’John Ramsay, who is responsible for The Famous Grouse, Cutty Sark and Lang’s Supreme, agrees. ‘I am not influenced by the traditional
geographical classifications. I go on weight and flavour; the character of the new make and the kind of wood it has been matured in – Spanish or American oak; sherry or bourbon or refill casks. I will group the product of a single distillery differently, according to how it has been matured.‘For example, I use Glengoyne in all my blends, but will specify all Spanish oak for Lang’s, three parts Spanish to two parts American for Grouse and one part Spanish to four parts American oak for Cutty Sark.In my view, regional styles are not helpful: Glengarioch can work like an Islay in a blend; Bunnahabhain like a Speyside, and so on. If I want to create a new blend with a certain character, I know which malts, in which woods, are most likely to deliver that character – and they might come from all over Scotland.’Some blenders use a simple ‘phenol and ester’ rating for their filling malts, not unlike that produced by John Lamond and Robin Tucek in The Malt Whisky File (1995), where they gave a ‘sweetness’ and ‘peatiness’ score out of ten for all the malts considered. ‘The ratings are a statement of fact,’ they say in their introduction, ‘a guide to help you find those malt whiskies which are most akin to your own taste. If, for example, you like a malt with a sweetness factor of seven and a peatiness factor of four, then those other malts which have a similar rating should be of interest to your palate.’There have been other attempts by the Scotch whisky industry and by liquor retailers to provide simple flavour guides for perplexed consumers. None of them have worked, and many consumers remain confused. For example, four years ago United Distillers embarked upon a complex research programme, codenamed Project Huxley, with the laudible task of coming up with a simple way of classifying both malt and blended whiskies. If memory serves they measured what they termed ‘intensity’ on an A-J scale, categorising whiskies as ‘light’, ‘medium’ and ‘full’. ‘Intensity’ was defined as the ‘overall drinking experience, from the enjoyment of the colour of the whisky to its aftertaste’, in spite of the fact that it has a somewhat different meaning to sensory scientists. This did not solve the problem.Clearly, classification by character, style or flavour is more useful to the consumer than mere geographical grouping. The most exhaustive attempt to do this was made last year by Dr David Wishart, a designer of statistical software, using a statistical method known as ‘cluster analysis’ to classify malt whiskies. Although originally developed for studies in biological taxonomy, cluster analysis can also be used for market analysis. Dr Wishart’s lassification is provisional and on-going. He is keen to have your comments on his findings to date, so please let us know your views and we will pass them on to him.How it works is this. Dr Wishart analysed the descriptive terms used in eight current books to describe 85 readily available single malts in proprietory bottlings at around ten years old. A vocabulary of some 800 aromatic and taste descriptors was compiled. These words were then bundled into a number of flavour/aroma groups: sweet, peaty, smoky, medicinal, honeyed, spicy, sherried, nutty, cereally, fruity and floral. Each of the 85 malts was ‘consensus coded’ (2 where a majority of authors agreed, 1 where a minority agreed, 0 otherwise) according to the number of times a descriptor was applied to it. Using his Clustan software, Dr Wishart then classified the 85 malts into ‘clusters’, each having broadly similar taste characteristics.The result is what’s called a ‘hierarchical classification tree’ in which the 85 malts have been ordered and classed into a kind of taxonomy of malt whisky based on their flavours and aromas. Dr Wishart then examined – somewhat arbitrarily – the division of this tree into ten groups of whiskies plus one singleton.
Although you may be surprised to find, for example, Knockando and Glen Grant clustered with The Macallan and Springbank, or Glenkinchie lumped with Highland Park, the methodology is interesting and the findings potentially of great value to the consumer. But to obtain more meaningful clusters, the language of whisky tasting must be more rigorous, the descriptors more narrowly defined, more analytical. The ideal body to do this would be the Scotch Whisky Research Institute. But since most of us, including this magazine’s distinguished Noses, do not suppress personal preferences and subjective assessment, the most meaningful clusters are personal, based upon our own experience. So why not set about producing your own clusters, based upon your own tasting notes
geographical classifications. I go on weight and flavour; the character of the new make and the kind of wood it has been matured in – Spanish or American oak; sherry or bourbon or refill casks. I will group the product of a single distillery differently, according to how it has been matured.‘For example, I use Glengoyne in all my blends, but will specify all Spanish oak for Lang’s, three parts Spanish to two parts American for Grouse and one part Spanish to four parts American oak for Cutty Sark.In my view, regional styles are not helpful: Glengarioch can work like an Islay in a blend; Bunnahabhain like a Speyside, and so on. If I want to create a new blend with a certain character, I know which malts, in which woods, are most likely to deliver that character – and they might come from all over Scotland.’Some blenders use a simple ‘phenol and ester’ rating for their filling malts, not unlike that produced by John Lamond and Robin Tucek in The Malt Whisky File (1995), where they gave a ‘sweetness’ and ‘peatiness’ score out of ten for all the malts considered. ‘The ratings are a statement of fact,’ they say in their introduction, ‘a guide to help you find those malt whiskies which are most akin to your own taste. If, for example, you like a malt with a sweetness factor of seven and a peatiness factor of four, then those other malts which have a similar rating should be of interest to your palate.’There have been other attempts by the Scotch whisky industry and by liquor retailers to provide simple flavour guides for perplexed consumers. None of them have worked, and many consumers remain confused. For example, four years ago United Distillers embarked upon a complex research programme, codenamed Project Huxley, with the laudible task of coming up with a simple way of classifying both malt and blended whiskies. If memory serves they measured what they termed ‘intensity’ on an A-J scale, categorising whiskies as ‘light’, ‘medium’ and ‘full’. ‘Intensity’ was defined as the ‘overall drinking experience, from the enjoyment of the colour of the whisky to its aftertaste’, in spite of the fact that it has a somewhat different meaning to sensory scientists. This did not solve the problem.Clearly, classification by character, style or flavour is more useful to the consumer than mere geographical grouping. The most exhaustive attempt to do this was made last year by Dr David Wishart, a designer of statistical software, using a statistical method known as ‘cluster analysis’ to classify malt whiskies. Although originally developed for studies in biological taxonomy, cluster analysis can also be used for market analysis. Dr Wishart’s lassification is provisional and on-going. He is keen to have your comments on his findings to date, so please let us know your views and we will pass them on to him.How it works is this. Dr Wishart analysed the descriptive terms used in eight current books to describe 85 readily available single malts in proprietory bottlings at around ten years old. A vocabulary of some 800 aromatic and taste descriptors was compiled. These words were then bundled into a number of flavour/aroma groups: sweet, peaty, smoky, medicinal, honeyed, spicy, sherried, nutty, cereally, fruity and floral. Each of the 85 malts was ‘consensus coded’ (2 where a majority of authors agreed, 1 where a minority agreed, 0 otherwise) according to the number of times a descriptor was applied to it. Using his Clustan software, Dr Wishart then classified the 85 malts into ‘clusters’, each having broadly similar taste characteristics.The result is what’s called a ‘hierarchical classification tree’ in which the 85 malts have been ordered and classed into a kind of taxonomy of malt whisky based on their flavours and aromas. Dr Wishart then examined – somewhat arbitrarily – the division of this tree into ten groups of whiskies plus one singleton.
Although you may be surprised to find, for example, Knockando and Glen Grant clustered with The Macallan and Springbank, or Glenkinchie lumped with Highland Park, the methodology is interesting and the findings potentially of great value to the consumer. But to obtain more meaningful clusters, the language of whisky tasting must be more rigorous, the descriptors more narrowly defined, more analytical. The ideal body to do this would be the Scotch Whisky Research Institute. But since most of us, including this magazine’s distinguished Noses, do not suppress personal preferences and subjective assessment, the most meaningful clusters are personal, based upon our own experience. So why not set about producing your own clusters, based upon your own tasting notes
