Myth 1: Ralph Harwood invented porter as a substitute for three-threads

No he didn’t, and no it wasn’t.

In 1802 one of the most influential articles in the history of beer was published. It appeared in a guidebook called The Picture of London, written by John Feltham, and the myths and inaccuracies Feltham recorded about the birth of porter, then by far the most popular drink in London, have been repeated continuously ever since, frequently in almost identical sentences to the ones Feltham used.

It was Feltham who first linked porter to an attempt to replicate a beer called three-threads, which he said consisted of a third of ale, a third of beer and a third of “twopenny” (strong pale ale); Feltham claimed that about 1730 “a brewer by the name of Harwood conceived the idea of making a liquor which should partake of the united flavours of ale, beer and twopenny. He did so and called it Entire or Entire-butt, meaning that it was served entirely from one cask; and as it was a very hearty nourishing liquor it was very suitable for porters and other working people. Hence it obtained the name porter.”

No previous writer in the 80-plus years since porter had first appeared had ever said the drink was an attempt to imitate another, mixed beer. Nor, apart from one brief attempt in 1788, almost 70 years after the event, had anyone before claimed Harwood of Shoreditch as the inventor of porter. You might think that, with a beer that became famous around the world, and brought fortunes to families including the Whitbreads, the Barclays and Perkinses, the Trumans, Hanburys and Buxtons and the Guinnesses, if the story about Harwood and three-threads were true, someone else might have written it down before Feltham.

There was a drink called three-threads, which is mentioned in a rhyming “good pub guide” of around 1718 called the Vade Mecum for Malt-Worms (a malt-worm being a slang expression for a beer drinker), but there is no contemporary evidence to show that it was made by mixing beers from three casks, that porter was a replacement for it, or that Ralph Harwood, who was indeed a brewer, in Shoreditch, East London, invented it.

None of this bothered later writers, who copied out Feltham’s story without attempting to do any investigation into the veracity of his claims. The Penny Magazine for March 1841, for example, contains an almost identical account of porter’s origins to Feltham’s 38 years earlier, while Richard Valpy French, in Nineteen Centuries of Drink in England, published in 1884, offered an easily-recognisable paraphrase of Feltham in his passage about the birth of porter. Another 25 years on, in 1909, and Frederick Hackwood’s Inns, Ales and Drinking Customs of Old England was still using whole sentences originally written by Feltham in his own account of the arrival of porter a century before (though Hackwood was wildly wrong with the year porter first arrived, suggesting it was “about 1750”)

A different version of the start of porter was given by a brewer called John Tuck, author of the Private Brewer’s Guide to the Art of Brewing Ale and Porter, published 1822, Tuck said that around the time of Queen Anne, early in the 18th century, London’s brewers. who sold a “heavy and glutinous” brown beer, started to come under pressure from the brewers of paler beers, which were popular with the country gentry now buying themselves houses in the capital. About 1720., Tuck said, London’s brewers brought out an “improved” brown beer “started, well hopped, into butts, and … kept a considerable time to grow mellow.” This, he said was the “intire butt beer” that caught on with the working, or portering classes, and became known as porter. No Harwood – no three-threads.

37 thoughts on “Myth 1: Ralph Harwood invented porter as a substitute for three-threads

  1. Mmm, it’s the Meantime Brewery boys – I like their porter a great deal, but I’m not so keen on every aspect of their history – for example

    hops of the day… Varieties were unknown, outside those of strict geography

    No – there were several named varieties, such as white bine, red bine and grape; and

    London brewers could not malt pale malts themselves until coke was introduced, as coal smoke was a problem in the capital”

    London brewers got most of their malt from Hertfordshire, which made both brown and white malts, and quite a lot of the rest from Norfolk and Berkshire, so smoke wasn’t an issue.

  2. […] We’re not talking household appliances with a “frappe” setting here.  Much like wine, scotch or cognac, mixing of beer of varying character  is vital to some styles (such as Belgium’s lambic). Porter, it is said, descends directly from “three-threads,” which three types of English ale blended together(although author and beer historian Martyn Cornell offers compelling evidence to dispute this). […]

  3. Hmmm, as an academic, I would not consider you rebuttal “definitive.” Not saying it isn’t true, but it is just 2 writers of beer with different historical perspectives. If you have any, more definitive proof, I would love to see it.

    I suspect that neither is 100% accurate, and both are probably their versions of myths that they had either grown up with, or heard at some point.

    As an ethnomusicologist, and beer teacher, I find that since no one really cared about when things were created 200 years ago or more, there is no real way to prove who is right. Music is very much the same when researching when a certain “style” or instrument was created.

    Could it be that more than one brewer invented Porter at about the same time? Could it be that an unnamed person created it, and the others just copied it without any thought of credit?

    There are lots of unanswered questions that may never be answered!

    1. Jon, there is no evidence AT ALL for the Harwood story except for one comment 60-plus years after the event and another 80 years after the event, which is provably wrong in its statement about three-threads. No commentator AT ALL between 1720 and 1760 mentions Harwood in connection with the origins of porter. And the “single hero” version of history is pretty much discredited anyway. What more do you want?

      1. Martyn,
        I am not doubting you, but if you can give me a resource that shows that (i.e. those books or articles you reference between 1720 and 1760), I will do my best to break the myth one beer class at a time!

        1. I can’t give you references to things that don’t exist. What I’m saying is that there is no book or article written between 1720 and 1760 that says Ralph or James Harwood, or anybody else, invented porter.

  4. Did you read Beer and Skittles The author Richard Boston cites H Jackson writing in “An essay on Bread” 1758. Porter first brewed by Harwood in Shoreditch (London) 1722. Porter being the beer of the populace.

    1. No, he doesn’t. I’m afraid you’re misreading Richard Boston. He quotes Henry Jackson about the “universal Cordial of the Populace”, and then Boston goes into his own words about the “Universal Cordial” being first brewed by Ralph Harwood, which he has taken either from Feltham’s narrative or one of the numerous writers who followed Feltham’s account: Jackson doesn’t mention Harwood at all.

    1. Look, this is the null hypothesis problem again. The null hypothesis is that Harwood did NOT invent porter. That’s my position. I don’t have to provide proof of that. If you disagree, YOU have to provide proof that he DID invent porter. Good luck with that. As I’ve just pointed out to David Rogers, Henry Jackson doesn’t mention Harwood as the originator: the first claim that porter was invented by Harwood did not come until 1788, that is, 70 or so years after the event. Since porter, as Jackson says, was easily the most popular beer in London by the late 1750s, and was already making the big porter brewers extremely wealthy (but not the Harwoods, who remained comparatively small operators), do you not think that SOMEONE in the decades between the 1720s and the 1780s would have mentioned Harwood’s name as the inventor of this hugely popular beer, if he had actually been its inventor? The fact that no one before 1788 is mentioned as porter’s inventor is very strong evidence, to me, that no one “invented” it, and instead, as argued convincingly by Dr James Sumner, it was developed out of the existing style of London brown beer.

  5. In the more than four ages before Porter arose in England (around 1720) people drank Poorter in the Low Countries. And it must have been really popular; it was the local beer, good for the local economy and tax measures were in place to support it. Sources point out that the meaning of the word poorter even changed to beer if the context implied this. One of these sources is from the Dutch province of Zeeland (around 1550). It is well known that around there one sailed to England, doubtlessly with Poorter onboard.

  6. Hello,

    As a graduate student in history writing on this subject I can perhaps shed a bit of light on the subject. There is no direct evidence that Ralph Harwood created the first porter beer in 1722, or 1730, or 1750. Perhaps the definitive account of the rise of porter brewing in the 18th century, Peter Mathias’s “The Brewing Industry in England, 1700-1830”, mentions Harwood as a possible creator of porter; however, the first definitive evidence of a beer called porter is traced to 1726 by Mathias. The first mention in the brewing industry traced by Mathias is a mention in “The London and Country Brewer” in 1738. So, no smoking gun proving Harwood as the creator.

    What I do have to disagree with is the characterization of “three-threads” where you state that “there is no contemporary evidence to show that it was made by mixing beers from three casks”. There are a host of documents that prove the practice of mixing beers in the 17th and 18th centuries. As early as 1663 Parliament had banned the practice of mixing different worts or brews and enacted further bans on the practice in 1670-1, 1689, 1696-97, and 1702. Parliament tried to stop the practice because brewers were making what a 1713 account by E. Denneston called “a thick liquid that very much resembled molasses, of such a quality, that in half and hour’s time it would make small beer strong.” That thick, heavily-hopped, and high alcohol beer-concentrate was a feature of the London brewing industry by the 1690s and was popular because it paid the same tax rate as other “strong” beers. When it was diluted with small ale or beer or another strong beer it could create a beer that had the same alcohol content as a “strong” beer or ale for a cheaper price. So the story that the mixing of “three-threads”, which when mixed was a dark beer, heavily hopped, and relatively high in alcohol content (as 18th century porters were), was most likely the result of the taste of Londoners for “three-threads” that porter got its start. The reason porter arose as an alternative to “three-threads” in the 1720s and 1730s was the government’s shift away from taxing only the end product (beer or ale) to the taxation of both the end product and the raw materials of beer- malt and hops. Once malt became taxed the advantage of “three-threads” fell apart as the more malt a brewer used the more tax they would pay, which meant the tax advantage that “three-threads” once enjoyed was no longer. Londoners did not lose the taste for “three-threads”, however, and a beer that replicated its flavor, porter, eventually became the beer of the 18th century.

    John Krenzke
    Ph.D. Candidate- Loyola University Chicago

    1. Thank you for pointing me to Denneston’s A Scheme for Advancing and Improving the Ancient and Noble Revenue of Excise Upon Beer, Ale and Other Branches, which I hadn’t come across, and which certainly throws important and fascinating light on Three-threads and mixed beers. But what Denneston says is that this mixing of super-strong beer, upon which had only been paid the same tax as “ordinary” strong beer, with lower-taxed small beer, to make “ordinary” strong beer, was a fraudulent practice to avoid tax, which would get the brewer and retailer fined if discovered. He certainly makes no link at all between three-threads and porter, and neither does anyone else, certainly not Obadiah Poundage, until Feltham in the early 19th century.

      There simply is no evidence that porter sprang from a desire to replace three-threads. And Denneston also indicates or implies that Three-threads was made from two casks, not three: the whole basis of Feltham’s story is that porter was supposedly invented to stop the faff of mixing three different casks together. In addition, three-threads continued to be mentioned through to the middle of the 18th century, long after there was any benefit to brewing super-strong beer and mixing it down to defraud the revenue. So it looks as if the taste for three-threads was (1) independent of any fraudulent aspect of the beer mixing and (2) still continuing even after the rise of porter. I still see no evidence for a claim that porter replicated three-threads: indeed, Poundage says porter was simply the ordinary London brown beer improved, with the London brewers hopping it more, lengthening the storage times and improving the ways they stored it. Have you seen the piece I wrote here?

  7. Martyn,

    Thanks for the reply. I can agree with your assessment that there is no definitive link between three-threads and porter. Even Mathias makes the point that the language describing the brewing of porter is ambiguous throughout the early and mid-18th century with brewers who were known to be brewing porter like Whitbread did not refer to it as so until the mid-18th century- 1760 in Whitbread’s case.

    What I can say, however, is that three-threads is referred to specifically as a mixture of brews in a manuscript from the British Library titled “An account of the the losse in the excise on beer and and ale for several yeares last paste, with meanes proposed for advancing that revenue”. The document is undated, but can be dated between 1698 (the last year of excise data the author used) and 1713 (when the rate of the Malt Duty used by the author in computations changed). In it the author, who was probably an Excise or Treasury official based upon the access they had to Excise data from 1683 to 1698, makes plain that the beer concentrate I mentioned previously was being mixed with other beers to create three-threads. Here are the pertinent passages from my own transcription of the original in the British Library:

    “This Duty of Excise as now laid is very unequally proportioned in respect of the several qualities of Beer & Ale brewed of late more than formerly. As for Example, very Small Beer of 3s. per Barrell pays 1s. 3d. Duty, Common Strong Beer and Ale pays 4s. 9d. per Barr, being the Strong from four Bushells of mault & sold for 18s. a Barrell. Very Strong Beer or ale the Barrell being the Strong from 8 Bushells pays on 4s. 9d. and yet sells for £3 so that tis Demonstrable that the Duty is unequally proportioned which has begot a kind of trade of Defrauding. For a Barrell of Strong Beer brewed from 8 Bushels of Mault pays 4s. 9d. and 2 Barr of Small (?) Drinke pays 2s. 6d. which together makes 7s. 6d. Now mix the 1/3 of Strong & 2/3 parte of Small together and it makes 3 Barr of Common Strong Beer the duty whereof ought to bee 14s. 3d. But by reason of the strength of the Barr from 8 Bushels thus mix’t the King looses 7s. out of 14s. 6d. of his Duty.”

    “I say wee have good Cause from that (?) to affirme, that a very great parte of the Losse aforesaid had arisen and does arise from the Frauds of the Brewers, etc. In like manner Consider which (?) That a Barrell of very strong Drinke of £3 (?) per Barr price, pays no more Duty as the Law now is , than a Barrell of Common Standard Strong Drinke of 18s. per Barrell. By Brewing and Vending whereof tis safely (?) to Contrive what Duty may bee saved, That the Brewers and traders in Drinks have not omitted (of late especially) to make the best use of this advantage, very strong Drinke being now Commonly a parte of the Brewers Guiles, and the whole of many who Brew nothing else. That the Consumption of it is everywhere, which you have under several odd names, as Two Threades, 3 Threades, Stout or according as the Drinker will have it in price, from 3d. to 9d. the quarte. That the notion thereof and Profitt thereby has been of late & now is generally knowne, And that the traders therein have turned themselves more and more to the practice of Brewing it , & skill doe as they can prevaile upon their Customers to vend it, when wee have bestowed a few Thoughts on those particulars wee may Conclude without being Stopt or Interrupting ourselves Concerning the very small sums Charged to have been the increase by means of Strong Drinke as before recited.” (Lansdowne MS 829/6- Fo. 6, British Library, London)

    What is clear from that passage is that three-threads was one of several mixtures of beer common in the period (1698-1713) that were being drunk to defraud the government of revenue through the excise on beer. Furthermore, the price data given by the author for the mixtures demonstrates that is was up to the drinker and vendor to create a mixture known as three-threads based upon how much they wished to pay. There would be no reason to pay 9d. a quart for a mixture unless it was qualitatively better in the eyes of the drinker, which could be achieved by using a larger proportion of beer concentrate (to provide more alcohol) or by mixing different styles of beer (to provide a specific flavor that the drinker desired). Now that could come from mixing beers from just two casks, but it could equally well have come from mixing beers from three or even more casks. While I agree the name three-threads may be a red herring in that it was always made from beer from three casks, I cannot agree that “the taste for three-threads was (1) independent of any fraudulent aspect of the beer mixing”. While the taste for three-threads in the middle of the 18th century quite probably was independent of any fraudulent aspect of defrauding the revenue it is quite clear that at one point (1698-1713) the taste for three-threads was undoubtedly shaped by a desire to defraud the excise.

    John Krenzke
    Ph.D. Candidate
    Loyola University Chicago

    1. Fascinating. A few comments:

      What is clear from that passage is that three-threads was one of several mixtures of beer common in the period (1698-1713) that were being drunk to defraud the government of revenue through the excise on beer.

      I think you have a good argument for saying that three-threads was being sold to drinkers as an attempt to defraud the revenue, being, if this narrative is correct, a mixture of very strong beer and small beer sold for the same price (or less) as strong beer but which had paid less tax than strong beer. But the evidence is less secure that it was being drunk to defraud the revenue, unless the drinker was getting something as good as strong ale for a lower price.

      Obadiah Poundage said in 1761 that three-threads cost 3d a pot/quart, and if it was as strong as strong beer, which he said was 4d a pot, obviously that would be an incentive to the drinker to buy it. But I don’t know of any evidence that says three-threads WAS as strong as strong beer. John Tuck’s The Private Brewer’s Guide to the Art of Brewing Ale and Porter, published in 1822, said three-threads was something different again, “a mixture of stale, mild and pale” – and also said it sold for four pence a quart. That would make it as expensive as strong beer. But it’s not, there, a mixture of super-strong and small. And Tuck, of course, was writing more than a century after the event.

      Apart from your references, the other contemporary descriptions of three-threads don’t call it a mixture of small and super-strong: the Dictionary of the Canting Crew by “BE”, published around 1697/1699, called three-threads “half common Ale and the rest Stout or Double Beer” , a definition repeated in Nathan Bailey’s Universal Etymological English Dictionary of 1737, which again said that three-threads was “half common Ale and half Stout or double Beer”.

      Incidentally, I meant to say that the first mention of porter is actually earlier than 1726: a pamphlet by the political journalist and poet Nicholas Amhurst dated May 22 1721 talks about dining at a cook’s shop “upon beef, cabbage and porter”.

      1. I’m coming to this discussion, which itself postdated Martyn’s post by some years, a little late. It would be good to see it continue, perhaps Martyn will be minded to revisit the vexed question of three (and other) threads and entire/porter.

        I have some issues with the term beer concentrate since I doubt that the strong beers made for mixing were as strong as that term might suggest to some (by the way Edward Denneston’s essay was cited in Peter Clark’s English Alehouse book and also in Peter Mathias’s landmark study). It is difficult to make a beer much over 11-12% ABV since the yeasts get killed off in the process. Given chancy early technologies, I doubt beers stronger than that were brewed unless brandy or other spirits were added – a possibility to be sure. Still, 11-12% is pretty strong and mixed with weaker beer probably did result in some beer being sold as strong beer.

        There is no clear evidence three threads was replaced by porter but I think it did happen based on much circumstantial evidence including that beer mixing and three threads are referred to in Poundage’s published letter (in the London Chronicle initially). The reference to “threads three” in the malt worm’s guide which Martyn has written about earlier clearly were referencing a mixed product (although one doesn’t know how the mix was produced).

        If three threads died out due to the advent of the tax on malt, something had to replace it. It was porter, namely a beer aged some months to be “racy” but not too stale (Poundage said), “mellow”, just right. I think the London three threads must have had that character. If a mixture of strong beer and sweetish ale (per the Bailey definition), it would have been bitter sweet, and perhaps a touch vinous where the strong beer was long-aged as beer then by definition usually was (except for small beer) – the heavier hopping of beer was to preserve it for a time, that is why it was beer. And Poundage/Combrune or whoever he was says entire butt beer had all these characteristics. Porter could not have come out of nowhere or evolved independent from the mixtures known to be sold in the early 1700’s. True, mixing continued after but it seems more an adjustment to the palate – “this entire doesn’t please governor, add a dash of mild beer please”. The brewers did it too, we know this e.g. from George Watkins’s brewing text of circa-1760. And finally the brewers evolved a short cut to use mostly mild beer with addition of a little strong stale or returns or whatever – back to the original method, but for different reasons (to save costs of laying away all that perfectly tuned porter, not taxes as such).

        Feltham may not have been right about 3 beers being mixed – probably it was only two as some have surmised from Deneston – but the essence of what he said was right IMO and consistent essentially with Poundage. Anyway sometimes three beers may have been mixed, that is possible, even the bruited ale, beer and twopenny.

        Where the term thread comes from is another question, one I’ve expressed thoughts on in the past, but it is not strictly relevant to the current discussion.


        P.S. I was glad to see the references to “poorter” by F. Ruis. I have always had a sneaking suspicion that the term and the beer have something to do with English porter. But it might, too, come from the looming industry, as I’ve suggested earlier again.

  8. Quick further postscript to say I do find Martyn’s position that Harwood did not invent porter persuasive. Poundage doesn’t mention James Harwood and that is an important point. But the real questions IMO were and remain, i) what exactly were two threads, three threads, five threads and six; ii) did porter emerge to take their place or at least the place of three threads, the presumed mean in strength.


  9. […] Sometimes it seems as though if a style of beer is to become popular then it needs a compelling back story, or a completely fabricated, mythological pile of codswallop. Whether it is the disgruntled citizens of Plzeň smashing up barrels of beer on the steps of the city hall in 1839, leading to the creation of the pale lager that now bears the city’s name, or IPA being invented with extra hops and alcohol to survive the journey to the mysterious east, it sometimes seems that tall tales and beer go hand in hand. Another beer with a myth of grand proportions is porter, that dark brew that according to legend was invented to replace a mix of beers known as ‘three threads’, a myth thoroughly debunked by Martyn. […]

  10. Yep…Martyn is right (probably); (a lifeboat of informed sanity in a sea of tsunami sized b!**$^!t. Keep us sane from the dross spewed forth from Eamon Holmes, Sky News, the anti drinks lobby & co)!!!!!

    I wish to make the point, my statement is a generalisation and not in relation to this topic…which I found very interesting (and thank you John Krenzke for your comments).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.