Date: November 2007

Column/Title: Berries, Wine and Chocolate (Some of the Time): Dry-Processed Coffees of Ethiopia and Yemen

Author: Kenneth Davids and Miguel Meza

------------------------------------------------------------------------------------------------------------------------------------

Dry-processed or "natural" coffees from Yemen and Ethiopia (those coffees dried inside the fruit rather than after the fruit has been removed) are the world's original coffees - and also, in their most recent incarnation, one of the newest trends in coffee. Unfortunately, most of the exotically distinctive coffees I reviewed this month with co-cupper Miguel Meza represent the new trend rather than the historic tradition. "Unfortunately" because many of us revere the traditional dry-processed coffees of Yemen and the Harrar region of eastern Ethiopia as living treasures, a direct link to coffee's first debut on the stage of world history.

However, of the thirty-five samples we sourced for this month's review, few of the more generic Yemen and Harrar coffees finished with ratings over 85. Many finished with very low ratings, ratings that push them down into the numeric nether regions normally inhabited by commercial canned coffees. Keep in mind that our cuppings are performed blind - neither I nor Miguel had anything other than a three-digit number to identify the samples until after we assigned our ratings and crafted our basic descriptors. We were most definitely not influenced by the names on the coffees.

Let's hope that this apparent deterioration of the great traditional dry-processed coffees of Yemen and Harrar is a temporary aberration. But for this year, at least, the drop-off is dramatic. In 2000 a traditional Harrar coffee scored 93 in a Coffee Review panel cupping, the first coffee in the history of panel cuppings at Coffee Review to score over 90. The highest-rated traditional Harrar this year scored 84. Factoring in the ratings inflation that has occurred across the industry in the intervening years, I would estimate that the 2000 Harrar rated as many as 12 points higher than this year's highest rated Harrar.

In a companion article to this one, Ethiopian Naturals: Present and Future, Miguel Meza provides some reasons for the hopefully temporary decline of traditional Ethiopia dry-processed coffees. Certainly one of the main problems with both the Harrar and Yemen samples we cupped this month was fading - they tasted woody and old. Perhaps many were last year's crop. As Miguel points out, getting green coffees shipped out of Ethiopia seems particularly difficult these days. Coffees sit in warehouses losing freshness and vitality for weeks stretching into months before they are finally shipped and eventually roasted.

Ironically, at the same moment that dry-processed coffees from Harrar in eastern Ethiopia have been enduring a couple of bad years, in southern Ethiopia (Yirgacheffe, Sidamo) coffee producers and exporters have begun collaborating with buyers to develop new versions of dry-processed coffees that take advantage of the more sophisticated fruit-sorting and drying methods originally introduced for washed coffees. What you get with these "new naturals," in other words, are the more rigorous sorting and drying procedures normally reserved for washed coffees applied to the ancient procedures of dry processing.

The result has been some exceptional coffees. One of the most impressive coffees Starbucks offered in recent years was its Shirkina Sun-Dried Sidamo, a recently developed (or encouraged) dry-processed coffee from southern Ethiopia. (This offering has been temporarily suspended by Starbucks, apparently owing to fall-out from a trade-marking dispute with the Ethiopian government.)

Recent efforts by Joseph Brodsky of Novo Coffee and his Ninety-Plus-Coffees import program have made coffees of this style available to boutique roasters. These are Ethiopian dry-processed coffees apparently meticulously sorted, rigorously protected from moisture during drying and subjected to careful grading and sorting after drying, giving us all the sweet fruit of the style without the musty pain - the lushly sweet, wildly fruity, often lightly fermented dry-processed profile without the downside of musty, mildewed sharpness or composty overripness.

But what about transport? What about the weeks and months sitting in warehouses waiting to be shipped? Still a problem, but one overcome in most of the over-90-rated coffees reviewed here by air-shipping. Many of the tiny lots of splendidly fresh coffees that top this month's ratings were shipped to North America by air, including the explosive Aricha Selection Seven, which netted four 90-or-over ratings, and the somewhat quieter Biloya Selection One, which attracted two.

So, based on this month's perhaps limited sampling, here we are: Either buy from a small pool of expensive, specially prepared, often air-shipped coffees that demonstrate this great traditional coffee profile at its finest, or buy from lots that are less expensive, more widely available, but in one way or another compromised by fading in the warehouse, mildly flawed sorting and drying procedures, or both.

For example, this month's Kickapoo Organic Ethiopia (89) I would guess was an impressive coffee when freshly hulled, but as we have it now it has become an agreeable but rather subdued and blurred version of the Ethiopia natural profile.

Of the six Yemens that turned up, the Batdorf & Bronson (reviewed here at 87) was the highest-rated. It is an interesting coffee, appealing to certain palates that enjoy the wilder style of Sulawesi or Sumatra coffee, for example, but given the musty tones (deriving either from flawed drying or long storage) that dampen and overlay the fruit notes, this profile cannot compare to the explosively sweet blueberry, wine and chocolate character of the meticulously sorted, carefully dried and expensively air-shipped small lots of Aricha Selection Seven and Biloya Selection One.

Differences among the four reviewed versions of Aricha Selection Seven probably relate mainly to differences in roasting, both in regard to "darkness" or degree of roast and roast profile (the pattern of application of heat and convection to the roast chamber). Without weighing casual readers with too much detail, it would seem from matching our ratings against machine-reading of bean color that the Aricha showed best when the roast was terminated at a quite "light" degree, well short of the second crack (the moment when the taste of the roast begins to dominate) with a relatively efficient, high-convection application of heat that encouraged a wide difference of roast color inside the bean.

Of course, as most readers know, what works to bring out the individuality of one coffee doesn't always work for another. In last month's review of Latin American competition winning coffees, for example, the two highest-rated of eight submitted versions of the competition-winning Guatemala San Jose Ocana appeared to show best when the roast was terminated at a point quite close to the second crack, darker than any of the current review's Aricha Sevens and much darker than the top-rated Barrington version.

Processing method alone, of course, does not in itself explain why this month's highest rated Ethiopia natural coffees are so explosively distinctive. Certainly a second major contributor is the botanical variety of the trees.

Both Ethiopia and Yemen are home to many, probably hundreds, of traditional cultivars of arabica that are planted nowhere else in the world. When I worked in Yemen many years ago, I found that every little group of valleys seemed to have its own set of varieties. Ethiopian coffee authorities have told me that in the Yirgacheffe district (the origin of all but one of this month's 90-and-over rated Ethiopia coffees); no outside varieties have ever been introduced into the region, meaning all Yirgacheffe coffee is produced from local varieties.

And wherever native Ethiopia or Yemen varieties have been planted outside the region - in India, for example (Selection 4, from Ethiopia), Panama (the famous Gesha, from Ethiopia), Hawaii (Moka, probably from Yemen) - they have brought with them a little to a lot of the floral and fruit character of Ethiopia and Yemen coffees.

Thus botanical variety and processing method together appear essential to producing the classic Ethiopian natural profile represented by this month's top-rated coffees. The relative importance of variety also is demonstrated by the interesting oddball in this month's reviews, a Panama coffee dry-processed (in Panama) using the best new Ethiopian methods (Flying Goat Coffee's Panama Los Lajones Natural, 89). Innovating farmer Graciano Cruz produced this coffee, which attempts to answer the question: What happens if you take fruit from trees of relatively neutral-tasting Central American varieties that are traditionally processed by the wet method and subject this same fruit to Ethiopian-style dry processing? I have heard signals from many places in the coffee world of efforts to develop new dry-processed versions of traditionally wet-processed coffees, but this is one of the first examples to reach us at Coffee Review.

It turns out that the Los Lajones Natural is a very attractive coffee, with some of the fruity, lightly fermented appeal of the Ethiopia dry-processed type, but still rather restrained by comparison to the lavish fruit, flowers, chocolate and wine bouquet of the most distinctive of the Ethiopia naturals reviewed here. Although higher growing elevation and not-yet-understood influences of soil and climate may have contributed to the differences between the Los Lajones and coffees like the Aricha Seven, certainly one of the main determinants driving the difference between them is botanical variety.

A Last Note: Dry-processing, Coffee Chemistry and Cup Profile

Over the centuries, coffee producers worldwide have tended to divide themselves into two camps: those who carefully wet-process the best of their coffees and those who dry-process various coffees that are already defined as second-rate - coffees of the robusta species, coffees rejected by the wet mills, lower-grown coffees.

Only Brazil and Yemen and the Harrar regions of eastern Ethiopia have preserved the practice of dry-processing all of their coffee, not just the rejected or already overripe fruit. And now Brazil is well on its way to abandoning the practice also, with producers increasingly choosing to process their ripe fruit by a compromise method called the pulped natural or semi-dry process.

Recent research has confirmed what cuppers already know: considerably more glucose and fructose develop in dry-processed coffee than in wet, causing properly dry-processed coffee to read as fundamentally sweeter in the cup. Unfortunately, processes other than glucose and fructose development can go on inside the slow-drying coffee fruit: sugars may ferment, for example (not all bad, since mild, clean ferment produces wine- and brandy-like notes that many coffee drinkers find attractive). But the ferment may also attract microorganisms, which can produce flavors ranging from mildly musty to sharply musty to medicinal to outright rotten. A really bad dry-processed coffee is a horror show of sensory insult.

Thus the trick to producing fine dry-processed coffee is to maximize the fructose, control the ferment without eliminating it entirely, and above all avoid encouraging the development of musty, medicinal and rotten flavors. The basic way to do this is to dry only sound, healthy, ripe-to-slightly-overripe fruit, promote even drying, and avoid moisture or rewetting the fruit during drying.

Interestingly, recent research by Dr. Dirk Selmar of the University of Braunschweig has found that, even if the drying is perfect, what Dr. Selmar characterizes as "drought stress" inside the partly germinating dry-processed coffee seeds encourages accumulation of y-amino butyric acid or GABA, a substance associated with bitterness and astringency.

This last explains, I think, why some of the finest and most perfectly processed of the coffees reviewed this month still show a slight tightness in cup and finish, a faint shadow note that cuppers may variously characterize as bitter, astringent or salty. If Dr. Selmar's research holds up, then I think that the cupping community needs to figure out how it wants to respond to this note: punish every Ethiopia natural that shows even a hint of it, or at very mild, often barely detectible levels treat it as a forgivable by-product of a processing method that can net otherwise explosively positive and distinctive fruit character.

Miguel is a leading member of a new generation of specialty coffee leaders, younger coffee professionals who typically began their careers as baristas before deepening their involvement by becoming roasters, blenders and green coffee buyers. Miguel is all three at his family's business, Paradise Roasters, founded in 2002. Miguel is particularly suited to co-cup for this set of November 2007 reviews of Ethiopia dry-processed coffees owing to his enthusiasm for the type and his deep engagement with it. In 2006 he participated as a judge in the ECAFE competition in Ethiopia and he brings a palate tuned to the type by cupping hundreds of examples since that visit.

------------------------------------------------------------------------------------------------------------------------------------
2007 The Coffee Review. All rights reserved.