On Sotheby’s 5th Edition

Tom Stevenson opens the pages behind his Sotheby’s Wine Encyclopedia, its perceived location in the geography of wine literature, its historical connection with the very first edition of Johnson’s The World Atlas of Wine, and highlights what’s new in the 5th Edition, which was published by DK on 31 October 2011.


I would never dream of writing anything as self-indulgent as this had Matt Stamp not specifically requested it, but once asked, how could I refuse?  Having just released latest edition of Sotheby’s Wine Encyclopedia, the invitation to elaborate on its revision process and major differences could not have arrived at a more opportune moment.  And what more appropriate and knowledgeable readership could there be than the Guild of Sommeliers, whose members are regular buyers of my book?

Revising Sotheby’s Wine Encyclopedia is a bit like painting the Forth Bridge.  Although there was no truth to that mythically endless task, the effort to keep any substantial reference work up to date is a never-ending job.  I was recently discussing this very topic with Ron Jackson, author of Wine Science – Principles and Applications, whose 4th edition is due out in 2014.  We were talking about how long it took to complete a revise and what drives him, and he told me ‘It is certainly not worth it in terms of money.  Suzanne [his late wife] once did a quick calculation that indicated that I was working for between 25 to 50 cents an hour.’  Thank goodness he loves his job because Wine Science is, as far as I am concerned, the best single volume work on the subject and I urge any sommelier who does not already own a copy to buy one.  It offers the same depth of knowledge as the equally excellent two-volume Handbook of Enology by Ribéreau-Gayon et al, but Wine Science covers a wider range of subjects (including wine law, wine assessment, wine and health, and more), the Contents for which lay readers in particular will find more intuitively named and more logically ordered, yet the degree to which he tackles his subject is second to none.  If anyone thinks my job is difficult, then Ron’s must be close to impossible.


The revision process

As soon as the first advance copy of Sotheby’s Wine Encyclopedia arrives hot off the press, I start scribbling on its pages, adding this, cutting that, and littering the text with questionmarks wherever I expect changes to occur in the coming years.  I then take out a marker and boldly highlight all the information that is definitely time-sensitive and must be updated as late as possible in schedule.  When friends happen to stop by and see me defacing a pristine copy, they are shocked.  They look at me in horror and ask how I could possibly inflict such wilful damage on any brand new book, let alone one of my own.  But it must be done and I do not stop my scribbling until the very end of the process because, if it’s important enough, I can always squeeze in more information at proof stage.  If it is exceptionally important, I have been known to delete existing text at proof stage to squeeze in new information. I have three years in which to produce a revision: the first year is consumed by travelling, tasting, researching; the second year my travelling is severely reduced as I update maps and create new ones; and the third year I lock myself in the office and write, restricting my tasting to incoming bottles to catch-up on vintages. Well, that’s the plan, but it seldom goes as smoothly as that.

The idea behind updating the cartography before writing new text is that the methodical process of placing new and altered appellations on a map lays down a foundation that gives structure to a large part of the research I must conduct.  Maps are essential.  As I write in the introduction to the 5th Edition ‘If there is one thing that can help us remain anchored while contemplating this ever-growing morass of vinous information, it is a sense of place for wine, and the only way to project that is with maps.  If readers can sit in an armchair, sip a delicious wine, look at a map and say “Ah, that’s where it comes from!” then I have done my job.’


The Origins of Sotheby’s Wine Encyclopedia

My job would not have been conceivable but for Hugh Johnson’s The World Atlas of Wine. When this seminal work was first published by Mitchell Beazley in 1971 it changed my life.  I used to drink wine before Johnson, but after The World Atlas of Wine, I began to think wine.  There is also a close connection between his book and mine.  If you have a first edition of The World Atlas of Wine, turn to the imprint page and you will see that the Editorial Director was Christopher Dorling, and the Art Director was Peter Kindersley.  Flushed with the success of their work, they formed their own publishing company, Dorling Kindersley (DK), in 1974 and immediately became wildly successful with highly illustrated reference works on many subjects, but mysteriously failed when it came to wine.  In 1985, when I was beginning to attract a reputation for opening up wine appreciation for everyday consumers by establishing the UK’s first annual shop-by-shop, shelf-by-shelf, wine-by-wine guide (The Sunday Telegraph Good Wine Guide), I was asked by DK whether I could devise a world wine reference to compete with The World Atlas of Wine.

I nearly talked my way out of the contract by making it very clear that no wine book could possibly beat The World Atlas of Wine because not only was the concept right and the execution perfect, it was also published at a unique time in the history of wine appreciation.  The 60s and 70s saw a boom in international air travel, when flying stopped being a luxury and opened up mass tourism.  For their annual vacation, Britain’s blue-collared workers forsook Brighton and Blackpool for the Mediterranean and whilst most were so happy to keep returning to the Costa del Sol year after year that it became known as the Costa del Fish & Chips, a significant proportion began exploring the rest of Spain, then Italy, France, Germany, Greece and Portugal. While supplies of Watney’s Red Barrel had to be shipped to the Costa del Sol to satisfy a Little England in the Sun mentality, the more adventurous holidaymakers were enthusiastically sampling local wines throughout the rest of Europe.  When they returned home, there was a thirst for wine knowledge and, voila, Hugh Johnson’s The World Atlas of Wine.  Happily for Mitchell Beazley, Americans were also starting to get inquisitive about wine, so sales soared there too.

Although we all have our favourite classics, with some dating back to the 19th century, it has to said that most wine books prior to Johnson’s Atlas were very dry affairs indeed.  The air travel boom had started in the 1960s, but the nearest that almost all wine books got to a sense of place before The World Atlas of Wine was a crude line drawing that claimed to be a map.  In the early 1970s, ordinary British wine drinkers eager to see where the wines they drank on holiday came from could see with their own eyes, thanks to Johnson, who led readers on an armchair journey to other wines they should seek out.  It was not just about the maps or, indeed, Johnson’s powerfully descriptive text, it was also the fact that coloured illustration leapt off every page. Before colour on page printing became widespread, the best that wine drinkers searching for information could hope for would be two or four multi-paged sections of photographic plates.  These were, more often than not, in black and white, and always completely detached from the text they related to.  Just to turn the pages of Johnson’s The World Atlas of Wine and see full colour photographs and wine labels on the same page as the text they refer to brings each region to life.  When thinking of this book some 40 years later I still see in my mind’s eye the photo a glass of gold-green wine in the cellar of a Chablis grower and from the first moment I set my eyes of that photograph, I knew how great Chablis should taste.  I didn’t need to be told, the picture said it all.

I make no apologies for harping on about another book, especially one that happens to be the most important work in wine literature, but as I told DK when they were dangling a contract in front of me, it was the timing of their old publication that would prevent me from devising a book that could be as successful.  In fact, I doubted that any wine book in the future could ever equal its success.  The most we could hope for back in 1985 would be to become the second-highest selling wine reference of a similar size and price, which thankfully they accepted.  As there was no way we could beat the detail of Johnson’s maps (something that holds to this day, unless it’s a single-subject wine book with an unusually high budget for illustration), the obvious option was to exploit their only limitation: coverage.  The more detailed the cartographers are, the less they are able to map, unless they have the luxury of an even greater number of volumes than the Encyclopaedia Britannica.  I told DK that my aim therefore would be to map every single appellation that appears on a wine bottle, official and unofficial; to describe the wines produced in each appellation, defining as many as possible; to recommend the best producers in each appellation, and the best specific wines they produce.  In other words, to cover far more ground in one direction (maps/appellations) and in far greater detail in the other (appellations/wines), keeping clear of the middle ground that Hugh Johnson had so comprehensively staked such an effective claim to.  By doing this, I hoped to produce a book that answered nine out of ten questions most readers are likely to ask.  DK gave me the green light and the first edition came out in 1988, since when it has notched up audited sales in excess of 762,000 copies (despite the claim in the book itself that it has sold “only” 600,000 copies, but that’s another story).  Whichever figure you take, it is still the second highest selling wine reference in £35/$50 cover price bracket.  However, even I have to admit that it’s a very poor second compared to the 4 million copies sold of Hugh Johnson’s The World Atlas of Wine …


The 5th Edition 

So what’s different?  It is now 736 pages as opposed to 664 pages in the 4th Edition and 4th Edition Revised.  At the front, there has been a lot of change and additions.  Joanna Benwell, an excellent editor on the 4th Edition, asked me if I would mind if she rearranged the text at the front of the book to form a more logical order and, almost without thought, I gave her the go ahead.  An author of a reference work can become too involved, editors are supposed to have a better overview of such books, and Joanna was one of the very best editors I had worked with.  I therefore trusted her judgement and as soon as I saw 4th Edition at proof stage in 2005, I realised just how right she had been.  Although a flow from one topic to another always existed in the text at the front of the book up to and including the 3rd Edition Revised (2001), the topics themselves did not follow the order of Factors that Affect Taste and Quality.  The 4th Edition was a big improvement, but as soon as I started working on the 5th Edition, I took Joanna’s logic to its ultimate conclusion, providing a two-page opener for and a brief explanation of Factors Affecting Taste and Quality, followed by all text in strict order of Location, Climate, Aspect, Soil, Viticulture and Vinification, and Grape Varieties.

Within these sections there are a number of changes.  In Location, for example, there are now three maps illustrating where the most important viticultural areas are located by Temperate Zones and Ocean Currents, as well as by Latitude.  Under Soil, the “Guide to Vineyard Soils” has had a number of entries added.  In the Viticulture and Vinification section has received the greatest amount of alteration and additional text, from the relatively minimal updating of mechanical harvesting and pressing, to minor extra items, such as the box on white-winemaking techniques covering everything from the use of enzymes to lees stirring, VA lift and wild yeast ferments, to major new components, such as a fully-fledged Sparkling Winemaking section with its own work flow-chart.  I wanted to move some of the sparkling wine text from Champagne to this section in 2005 (which probably ignited Joanna’s interest in the reconfiguring the order of the material at the front of the book, which in turn sparked of my continued rationalisation in the new edition), but was unable to so because it was too late in the production schedule to add any more pages at the front of the book.  Six years later I got my chance and the logic was undeniable.  General methods should be looked at in detail in the front of the book, not in the Champagne chapter, as the methods involved do not apply only to Champagne.  Only what is unique to Champagne should be in the Champagne chapter.  Also, I could not believe that there was no work flow-chart for Sparkling Wine. In a global wine reference written by anyone, it would be a major error, but in a global wine reference written by someone who specialises in Champagne, it is unforgivable!  But the biggest addition to the Viticulture and Vinification section is the Barrelmakers chapter featuring profiles of 34 of the world’s most important coopers.  This is a topic I hope to expand further on in future editions.

One of the two biggest changes found in the front of the book is in the Grape Varieties section.  There are plenty of small differences, such as the number of genera of Vitaceae having risen from 10 to 17, and the interesting reason why, but the big, big change is the 22 page “ABC of Grape Varieties”, with over 1,000 entries!  It had been my intention to list every grape variety and synonym mentioned throughout the encyclopedia in the “Glossary of Grape Varieties” since the 3rd Edition (1997), but it was such a mammoth task and there was always something else even more important to do.  However, with Jancis working on a new book about grape varieties, expanding the coverage in my encyclopedia became imperative.  When the first edition was published in 1988, Sotheby’s World Wine Encyclopedia (as it was originally called) received many excellent reviews, but even some of the most enthusiastic reviewers considered my “Glossary of Grape Varieties” to be little more than a copy of Jancis Robinson’s Vines, Grapes and Wines.  It was not, of course. Completely the opposite, actually.  It was in fact an expansion of a two-part feature I had written for Decanter called An ABC of Grape Varieties, which was published in 1981, five years before Jancis Robinson’s brilliant book.  Once bitten, twice shy, as they say.  I dread to think what reviewers might have thought had I left it until after Jancis’s new book to compile a glossary so massively larger and different than anything that went before, so it was the 5th Edition or never.  It is now so comprehensive (and will grow with each new edition) that it should answer nine out of every ten questions on grape varieties.  For the answer to that tenth question, you’ll have to wait for Jancis’s new book, which will, I am sure, outclass even Pierre Galet’s Dictionnaire Encyclopédique des Cépages.

I should also point out that the super-comprehensive “ABC of Grape Varieties” coincides with a complete review of the grape varieties listed in Factors Affecting Taste and Quality boxes throughout the encyclopedia, which should now all conform to the same style of using the local name first (be that a synonym, such as Pinot Beurot, or foreign-language, such as Pinot Grigio) followed by, if necessary, the generally accepted or primary name in brackets (which would be Pinot Gris in this case).

Wine Glasses is a new feature examining the shift in stemware from regional tradition to technical shapes and my assessments of some of the best individual glasses based on an exhausting test of the complete range of nine manufactures.  Please do not judge me too harshly if observant readers notice my failure to update the Mikasa brand name (now called Chef & Sommelier).

However, you do not have to be very observant to notice that the photographs of me sniffing, swirling and spitting in The Taste of Wine section have changed!  I have not looked the guy in the old photographs for some time and had asked DK to convert them to drawings, but they said they had some very similar shots of Joanna Simon, if I did not mind featuring another person in my book.  I have never been a me-person, so I had nothing against someone else in principle, and I like Joanna a lot, so it was not going to be a problem, and now that the book is published, I think she brightens up the pages!  Further on in this section, you will find mini-essays on the disadvantage of being a “supertaster”, how the debunked tongue taste map evolved and umami.

In the 5th Edition, the old Taste Chart that was at the rear of the book, has been moved to the front, where it has been expanded and updated, and is now called Tastes and Aromas, with the Fault Finding chart immediately following.  By far the largest totally new feature in the 5th edition is A Chronology of Wine, which is 10 pages and spans 500 million years, with historical titbits ranging from a non-flowering climbing vine that is speculated to be a distant relative of Vitis vinifera and existed when the Earth had only one landmass, the supercontinent known as Pangaea, to a cache of the oldest surviving Champagne recovered from a shipwreck in 2010.  As I write in my Introduction, ‘How do you get a meaningful history into an encyclopedia such as this without it doubling the size of the book and making for heavy reading?  Well, this chronology is my answer, and I hope readers find it as fascinating to dip into as I do.’

All in all, the front sections have doubled in size for the 5th Edition.  This has left less room for expansion of the geographic chapters in the book, which now comprise 591 pages as opposed to 570 pages in the 4th Edition, but by judicious cutting of dated and less important text, and a little tweaking of the design in places, there is much more change than the small increase in pagination would otherwise suggest.

It goes without saying that all the data in boxes and charts have been updated, as have all the mentions in running text (or should be, perhaps I should claim 99.9 per cent, just in case something slipped through).  There are lots of new odds and ends, such as the explanatory box on AOPs or PDOs, which many people in the European wine trade still do not understand (it is not the replacement of AO, DOC etc., but the replacement of VQPRD or QWSPR), but the myriad of new detail really is for you to find out as it would take a book the size of Sotheby’s Wine Encyclopedia to spell out. 

The same could be said for almost all of the maps, where even those that required only minimal updating have benefited enormously from improved design.  While some maps are so crammed with appellations that improved design will have a negligible impact, most maps have been transformed by more effective use of colour, boundaries, and placing as many appellation names (if that’s not tautological) on the maps with pointers, rather than leaving it for readers work out whether a name is an appellation or a place.  This has resulted in a clearer, more immediate understanding, where the location and name of appellations is more intuitive on the first glance at a map and the times that readers are forced to consult a long and complex list of colour-coded keys is reduced.  It’s not yet perfect, but we are much closer to it than we have ever been, which is essential where the whole raison d’être for the mapping is to pinpoint appellations, rather than illustrate the detailed topography within a part of an appellation.

For a taster of the detailed updating that could go unnoticed until a reader looks for something specific, I will summarise a few of the minor changes to the maps in the French chapters before moving on to more major changes to maps found elsewhere in the book.  The Médoc communal maps look different because I have disposed of the gratuitous miniature line drawings of selected châteaux, and updated the area under vine, which is now colour-coded to show the difference between Cru Classé vineyards and other AOC communal vineyards.  The traditional appellations of France probably change less than any other wine area in the world, so it would be difficult on first glance to see, for example, the single blue line on the Libournais District map that indicates the Côtes de Bordeaux appellation, which was introduced in 2009.  There are much bigger changes, however, to both Chablis maps.  Some of the maps for the tradition French wine areas have been transformed more in a visual way than by boundary changes or new appellations added.  Take the Côte de Nuits map, for example, where the outline boundaries have been swapped for coloured infills, and the two single-vineyard Bourgogne appellations have been pulled out and magnified in an inset map.  The effect on the Côte de Beaune map is even more dramatic.  The Côte Chalonnaise map has been enlarged to take in Bourgogne Couchois.  Beaujolais and Mâconnais both have additional maps magnifying where each district overlaps the other, which in Beaujolais enables me to give a more detailed map for the crus.  There is a lot of updating changes to both of these maps that cannot easily be discerned until you start looking for specific named villages that are allowed to add there names to Mâcon-Villages and Beaujolais-Villages appellations.  This modest level of change can be found throughout France, but I can honestly say that no one, not even the French authorities, have as complete and as up-to-date Vin de Pays maps (there are now 154 vin de pays appellations or 168 if the sub-regional and sub-zonal vin de pays are included).  I know this because of the different sources I had to consult in order to bring everything together.

The Italian maps are amongst the most radically altered in the entire book.  There was a point sometime this year (2011) when I could put my hand on my heart and swear that all the 351 DOCs and DOCGs are mapped, and every one of the IGTs, DOCs and DOCGs is profiled, but Italian appellations grow like weeds.  One of the aims for the next edition has to be to map all the IGTs, something that was impossible with the 5th Edition because, with an increase of 163 DOC and DOCG appellations mapped and with so many of these crowded together and overlapping multiple appellations, it was impossible to add a further layer of 118 overlapping IGT appellations.  The only solution will be to have a budget for a complete raft of additional IGT maps ready for the next edition.

The Spanish map has been fully updated to encapsulate all the current DO, DOCA and Vino de Pago appellations, while the Portuguese map includes all IPR and VR/IGP appellations as well as mainstream DOCs (with sub-appellations added to the Dão and Vinho Verde regions).  Austria has subtly changed (e.g., Donauland is now known as Wagram), but like many of the maps it also benefits from an improved colour scheme and better use of line boundaries to highlight the larger regions.  Using the same scheme the Swiss map is much easier to discern and the language-based divides have been updated according to the latest census.  The map of Hungary is a huge improvement, as is Romania, which took a huge effort, not to mention arm-twisting, to pin down all the wine areas. The Czech Republic and particularly Slovakia are much more detailed, as is The Black and Caspian Seas and The Western Balkans, even though Croatia and Slovenia have been hived off for their own short chapters.  The Greek map is brilliant, even if I say so myself!  I wanted to map all the PGIs (the equivalent of French vins de pays) as well as all the well-known appellations on one map, but there was absolutely no cartographical reference to be found anywhere.  I began pestering the Greek authorities prior to the 4th Edition, but the only way that they could provide the necessary reference material was to commission their own map, which did not become available until well into the schedule for the 5th Edition and I really think that apply our design to the official map that we have a much better, far clearer finished product.

At last I have managed to start the process of giving Israel and Lebanon chapters of their own, which, like Croatia and Slovenia, they have long deserved.  In all four instances there is much more in situ research, tasting and writing to be accomplished for expansion in future editions, but at least I now have detailed maps upon which to build.  The only solution for the rapidly evolving South African wine areas was to magnify the central Cape region and put the large number of small Wine Wards in a separate map.  However, after I finished the South African maps and was confident that all the latest appellations had been added and every one of the many changes in classification noted, I discovered that a vast new district called Sutherland-Karoo had just created (albeit currently little used), which I had to map at the last moment and force through on pages that had I had long since signed off.  Then I noticed a corner of the Paarl that had been included on the official map just months before had suddenly disappeared!  Was it missed off by accident?  I cannot take such things for granted, so I made enquiries.  It turned out that when Franschhoek was promoted from Wine Ward to Wine District, it physically cut-off this corner, which had long been a Nature Reserve and would never be cultivated.  They took the opportunity of updating their own map to remove this corner, although there are hundreds of areas that can never be planted and no one has, or ever would, remove them because the map would look ridiculous, but if I did not do likewise, it would be my Sotheby’s map that would look odd.  And just when I thought that the South African maps were finally done and dusted, I found out that Wine Ward of Lower Orange had just been changed to Central Orange River!  Frustrating at the time, but rewarding to see everything taken in now, even if they have since added.

North Africa is properly mapped for the first time, which is ironic considering that its wine production has never been lower.  Amazing to think that Algeria was once the world’s second-largest wine-producing country.  The maps of The Americas have received a lot of care and attention, with every single appellation including all AVAs existing at the time of writing mapped, although some will need to be profiled in the text for a later edition. If you have the previous 4th Edition, please make a comparison of The Americas, California, Mendocino (including extended area maps), Sonoma, Napa (including extended area maps), Central Coast North (including extended area maps), Pacific Northwest and particularly Atlantic Northeast. With Ontario taking in Prince Edward County and BC the Gulf Island, the Canadian maps have barely changed.  Lots of change in South America, even on the large map, which now has much more detail for Brazil, Uruguay, Bolivia, Peru and even Venezuela. Chile and Argentina are as complete as can be found anywhere and, hopefully, more intuitively mapped.  The main Australian map and those of New South Wales, Victoria, South Australia and Western Australia underwent their greatest revision for the 4th Edition, after the country’s Geographical Indication (GI) system was introduced, but they did not suit the old Sotheby’s map design.  Under the new design, these maps are far easier to read and they have taken in all the GIs created since that edition.  For some strange reason Tasmania does have any GI zones, even though its grapes fetch the highest average price of all Australian states, but Sotheby’s Wine Encyclopedia does not restrict its maps to official appellations.  If wine areas exist, we map them and it has to be said that a detailed map of Tasmania’s wine areas was long overdue, as was its own chapter, both issues of which have been resolved in the 5th Edition.  By comparison New Zealand has had a minor overhaul, mostly in the area of the increased number of sub-regions. With Asia gaining in importance, it was about time that China, India and Japan were spun-off into their own chapters with more detailed individual maps.

The penultimate major change is the Micropedia (formerly known as “The Glossary of Tasting and Technical Terms”), which has taken on a life of its own in recent editions, and in the 5th Edition, this reference section has increased by 50 per cent.  This is no longer a glossary of tasting and simple technical terms, but also a quick fire reference to much more technical terms (such as Argon and Atypical Ageing), historical references (such as Anthosmias and Ariusian), and obscure references (Bellême and Berce).  All these examples are just from one page (689), but they are representative of additional material added for the 5th Edition.  There are also mini-essays on various subjects, such as Altec, Cork, Enzymes, Fruit-bombs, Micro-Oxygenation, Minerality etc, which have been added to existing mini-essays (Délestage, Flying Winemakers, French Paradox, Health Benefits of Wine, High-Density Vines etc) and, of course, many of the existing entries have been expanded and updated (for example Corked was 15 lines in the 4th Edition, whereas Cork Taint or Corked Wines is now 36 lines).  The idea is not to clutter up the main text with these terms and their explanation, which would turn off novice readers, and to put them into one section where they can be easily found, which is why this section will continue to grow over future editions.

Last but not least, I have always believed that the larger the book the more comprehensive the Index must be for readers to find what they are looking for.  It must also be focused to be practical.  What point is there having 20 page reference for every mention in passing?  None: readers would stop using the index.  Highlighting main entries in bold is a waste of space because only the main entries should be indexed, whether there is just one or four or five. I have never had enough space for the indexer to follow all of my instructions, so any wasted space is better utilised indexing something else.  For the 5th Edition I managed to get the index expanded from 22.5 to 36 pages, yet nine-tenths of my indexing instructions could not be followed owing to a lack of space, which makes me suspect that I won’t be happy with the indexing until I can afford to use 100 pages or more.  As I wrote at the beginning, this is a never-ending job.