A Turn-of-the Century British Account of Selected 19th Century Belgian Brewing Methods


Current interest in low-gravity beers brings new relevance to some of the unorthodox practices of 19th century Belgian brewing.

By Randy Mosher (Brewing Techniques)

I recently ran across a couple of papers in some old copies of the Journal of the Institute of Brewing, the venerable London-based brewing organization. They were written many years apart by George M. Johnson, a British brewer who had spent some time observing the production of Belgian Witbiers and other low-gravity ales. In 1895, Johnson was writing largely out of horrified astonishment, but 21 years later, at the height of World War I, the British brewers who had assembled at the Imperial Hotel were only too eager to have some advice on how to make the best of low-gravity beer. The rationing imposed by the war made this interest a necessity, whether they liked it or not.

Johnson’s descriptions shed light on the origin and operation of the bizarre brewhouse practices in use in Belgium at that time. His writing provides a snapshot of a certain type of brewing carried on during that period, specifically low-gravity ales and various white beers. This article examines these beers and their production. The many other splendors of Belgian brewing — Lambics, Trappist and abbey beers, brown ales, and the other idiosyncratic beers we now enjoy from that country — are outside the scope of this article.


Around the turn of the century, 2700 breweries were in operation in Belgium. Of those, only 25 were brewing lagers. Another 200 were brewing Lambic, and 300 were brewing abbey-style ales; a good portion of the rest were making top-fermenting beers by the methods discussed below.

Many of these were unadorned ales. Contemporary documents mention a Liège Saison, with original gravities of 1.020–1.025 (4.8–6 °P), highly hopped with 7–10 lb per quarter (0.3–0.48 oz/lb, 6–12 oz/bbl), and 10–15% of the grist was spelt (a grain with characteristics of both wheat and barley). Contemporary Belgian brewing authority Roger Mussche  also mentions La Blonde van Vlaanderen, another style made with oatmeal, but with calcium hydroxide (calcium hydrate or slaked lime) added. Jean de Clerck  teases us with a listing of other intriguing names: Uytzet des Flandres, l’Orge d’Anvers, and Diest, all described as “thick and dark”; Hoegaerde, “with a very acid palate”; and Zoeg “of Tirlemont, which was very sweet.

Johnson describes two Witbier styles: a Louvain White, made with 65–75% air-dried malt and 25–35% unmalted oats (1.025–1.030 O.G., 6–7 °P); and a Louvain Peeterman, in which unmalted wheat replaced the oats and to which calcium hydroxide was added to the kettle to increase the color. In brewing the white, unboiled wort was added to the fermentor, introducing Lactobacillus and other noncultured organisms. By the 1950s, original gravities had increased to 1.044–1.046 (10.5–11 °P), up considerably from the remarkably low gravities of the previous era.

The common characteristics for all of these beers are low original gravity and very limited attenuation, between 50 and 66%, far below today’s 75–80% range. This was achieved, in part, by the use of a Byzantine mashing arrangement — a kind of a backwards decoction, in which cloudy, enzyme-rich wort is withdrawn from the mash and boiled for a time before being returned.


Brewery technology and state of repair varied widely from rustic farmhouse facilities to modern, hygienic, steam-powered showcases. Ignoring the more up-market abbey breweries, our chronicler seems to be morbidly fascinated with the more humble sort.

Speaking from a very high horse, Johnson begins with a prejudiced pronouncement about the philosophical underpinning for these brewing methods: the all-pervading cheapness of Belgians. Only this, he believes, can explain the rusty cast-iron brewing equipment, the undersized mash-tuns, the chronic state of filthiness. A Victorian to the core, he wrote, “… I am of the opinion that if many breweries are still dirty, it is rather because cleaning costs trouble than because it costs money, and that when you have said that you have got to the bottom of the repugnance to cleanliness of the great bulk of humanity.”

In Belgium, as elsewhere, the approach to brewing was in fact more a result of tax policy than of anything else. In 1822, the Dutch overlords of what is now Belgium imposed a duty on beer based on the full volume of the mash tun rather than on the weight of the grains or the quantity of beer produced. This duty led to a desire to produce the maximum amount of beer from the tun. Overfilling demanded thick mashes, which in turn, create very full-bodied beers. As Johnson put it, “For the space of sixty years (1822–1885), the best minds in the Belgian brewing world seem to have concentrated on the problem of getting a quart into a pint pot.”

The inspired geniuses of Belgian brewing used every imaginable trick in their quest to best the tax man. Malt was finely ground, which allowed it to pack more tightly in the tun, and was even sometimes stripped of its husk to prevent inert ingredients from usurping the precious tun space (wheat husks or rice hulls were added as soon as the turbid mash was removed). Unmalted wheat, very finely ground, was a common adjunct.

Mechanical contrivances were also brought to bear on the problem. At first, an ancient device known as a “brewer’s basket” was used to separate the liquid from the very thick mash. A stout woven basket was pushed down into the mash, and the wort would trickle through the porous sides of the basket, where it could be scooped out and slogged over to the turbid wort copper, also called a “chain copper.” A chain copper was a boiling kettle fitted with rotating chains designed to keep particles from settling on the floor of the kettle where they would be subject to scorching. The basket concept evolved into metal versions of the same thing and, ultimately, into a series of perforated disks, “much like a pair of cymbals clapped together.” The disks were attached to hollow revolving arms and rolled through the mash like a farmer’s disk while turbid wort was pumped out through the supporting framework. The revolving action had the unwanted effect of centrifuging the mash towards the outer edges of the tun. This was a problem until someone invented an Archimedes screw–like version of this increasingly infernal mashing machine, which served to heap up the mash in the center of the tun, allowing even more mash to be precariously perched above the tun. At the same time, mash tuns became much wider and shallower, allowing for a much greater volume of this mash pile relative to the taxed volume of the tun.

Eventually, brewers were allowed a second vessel for the purpose of mashing wheat and other unmalted grains. This was also taxed on volume, but the government felt it necessary to allow some untaxed headspace, which the brewers immediately filled with mash. These secondary tuns perpetuated the use of raw grains, a practice that continues to this day in certain Belgian styles such as Witbier. In 1948 (translation published in 1958), de Clerck described a two-tun mash procedure for Witbier similar to those discussed here, so apparently the process survived nearly to the demise of the Witbier style. Lambic brewing follows a similar scheme for different reasons (a starchy, unfermentable wort is needed for the unconventional fermentation process used).


Ingredients: Beers of those times were made of a number of grains, with barley malt constituting at least 50% of the grist. The barley was usually a six-row winter variety, planted in November and harvested in July. It was very lightly modified compared with British malts. Johnson mentions the steely ends of the malt, which were capable of introducing ungelatinized starch into the mash. Kilning was rapid, and the application of high heat to still-moist grain caused some diastase to be destroyed. Air-dried malt was commonly used in Witbier, resulting in a very pale color. Raw wheat, said to produce “very full beers,” was an ingredient with centuries of tradition behind it. Pierre Celis, who revived the Witbier style, has stated that malted wheat “just doesn’t have much of a wheat taste,” and he prefers to use unmalted wheat in his Celis White. Oats were limited mainly to the production of Witbiers, in 25–35% proportions, although today’s Witbiers use only 5% oats in the mash. One of the articles mentions rye as an occasional adjunct. Maize and rice were introduced to Belgian brewers only in the 1880s and were used in quantities up to 25%.

Copper sugars (brewers’ sugars that were added to the copper) were also used in some breweries, although they were expensive compared with grain. A chip-type glucose was preferred, because it contained a proportion of dextrins so important to the character of these thick, low-gravity beers. Our dextrose/corn sugar would be entirely inappropriate, but a high-dextrin brewer’s corn syrup would fill the bill. Chip glucose is sometimes available from British homebrew suppliers.

Mashing: Apparently, the specifics of mash procedure differed from brewery to brewery, but Johnson gives two generalized schemes, one for all-malt beers and the other suited to the co-mashing of unmalted grain.

All-malt beers. For all- (or mostly) malt beers, the following procedure was followed: The grain was sometimes mashed in cold but more commonly at about 108 °F (42 °C) and was held for 0.5 h as a protein/acidification rest. The mash was diluted at 0.6 qt/lb. Then, 0.4–0.8 qt/lb of water at 158–167 °F (70–75 °C) was underlet, raising the mash temperature to 122–131 °F (50–55 °C). At this point, the mast taps were opened, allowing the cloudy, enzymerich portion to run off. This turbid wort was transferred to the “chain copper.” The wort was heated rapidly to boiling, with only a short rest (15–30 min, I would guess) for starch conversion along the way.

During this time, water near the boiling point was added to the mash, bringing the temperature of the goods up to 158 °F (70 °C). Because most of the diastase enzymes were drained away along with the cloudy wort, conversion was relatively slow. After 45 min, taps were opened and the bright wort run off into the conventional boiling copper. The boiled turbid wort was then added back at either boiling temperature or, to ensure more complete conversion, at 176 °F (80 °C). Typically, the resting temperature of this stage was about 167 °F (75 °C), favoring the production of dextrins. Once the wort ran clear, it was pumped up to the copper, joining the clear wort from the earlier mash, which had been maintained at about 167 °F (75 °C), to facilitate conversion of any undigested starch still remaining in it by the infinitesimal amounts of enzymes still present in the later, clear wort.

Adjunct beers. For adjunct beers, the process was somewhat different and arranged so that most of the dextrin was extracted from the adjunct, the malt providing mostly highly fermentable maltose. Cold water and finely ground adjunct grain (1.9 qt/lb) were placed in the chain copper, with 5% malt added to prevent balling. Over a period of 90 min, the adjunct mash was gradually heated to boiling and boiled for 20–30 min. After 1 h, the malt was mashed in with hot water for a 30-min rest at 108 °F (42 °C) and then slowly raised to 158 °F (70 °C) by small additions of hot water. Meanwhile the boiled adjunct was cooled to 176 °F (80 °C) and added to the malt mash, bringing the whole thing up to 158–167 °F (70–75 °C).

At these high temperatures, the starch in the unmalted grain is preferentially converted into unfermentable dextrins. With flaked (pregelatinized) grains, the precooking may be eliminated; the flakes may be added directly to one-third of the malt mash that was transferred to the chain copper and the mixture quickly raised to 158 °F (70 °C).

Runoff was as for all-malt beers but took longer because of the lower proportion of husks. Runoff was often assisted by adding oat or rice chaff as a filtering material. Louvain white beer was given only a short boil, but many of these styles were boiled for a very long time — from 4 to 10 h — a practice that was viewed as increasing the beer’s body. It certainly would have increased the richness of flavor of such light beers, adding a good deal of caramelly character and color as a result of Maillard reactions caused by extended heating of the wort. Long boils would have also extracted significant amounts of tannic material from the hops, which would have helped preserve the beer and also would have produced considerable astringency, which may have helped balance the inherent sweetness of these poorly attenuated beers.

Fermentation: These beers were almost always fermented in barrels as opposed to tuns or other open vessels. Despite much experimentation, British-style open fermentors invariably produced flat-tasting beers, largely devoid of carbon dioxide. Johnson notes that he measured temperatures as high as 78 °F (26 °C) in casks, whereas tuns tended to be much cooler. The temperature difference would produce a definite flavor difference. Not all beers were fermented this warm; more typical temperatures were in the mid-60s °F (~18 °C).

Early yeasts were, of course, mixed cultures. Hansen and others isolated pure strains, which failed to produce acceptable beers at first. Eventually mixed strains of clean yeasts found favor. The yeasts used varied widely from region to region, and many unique and even bizarre varieties found strong local followings. The yeasts of Liège, whose beers were especially lightweight, dropped out after the beer became 50% attenuated. This is especially amazing if we consider that these beers started at only 1.020–1.030!

The yeast in Louvain Peeterman was even weirder, producing pineapple and other aromas. According to Johnson, “the hydrogen gas would sometimes flash with a blue light if a match were brought into contact with gas escaping from it.”


I have yet to experiment with any of these techniques, but I plan to this coming brewing season. Hopefully, these methods will be applicable to the small brewery I use. Certainly, there is a demand for flavorful beers that are low in alcohol, something we have not seen for a very long time. Perhaps these extreme techniques hold the key to such products.

All contents copyright 2019 by MoreFlavor Inc. All rights reserved. No part of this document or the related files may be reproduced or transmitted in any form, by any means (electronic, photocopying, recording, or otherwise) without the prior written permission of the publisher.