As described in my last post, I recently received the following message from M.V. Ramana.

Dear Amitav,

Greetings from Princeton. I am enjoying River of Smoke, but a question has popped into my head and it refuses to go away. So I thought I would share it with you. Among all the books of yours that I have read so far, this has got to be the one with the most descriptions of food. From parathas to dan-dan noodles to mutton dhansak to chai garam, there is quite a range of cuisines and dishes that enter your book. At your talk in Princeton last semester, you spoke a bit about how you traverse the twin-universes of being a journalist (and sticking scrupulously to fact) and of being a fiction-writer (where you can take some liberties). The question that comes to my mind is how much of these descriptions are fact and how much is made up? Was chai garam really drunk (invented?) by Parsis in Canton in the early 19th century? Was there a dish called dan dan noodles that was eaten by the Chinese in Canton during that period? You get the gist. Since I know that you usually back up your books with a lot of research, I suspect you have evidence for these. If so, I would be curious about what sort of sources you go to for these.

Thank you and warm regards,

Since Ramana is a physicist and a member of the Coalition for Peace and Nuclear Disarmament, these queries were, I must admit, somewhat unexpected. But in a subsequent mail Ramana explains: ‘You may be onto something by positing a connection between my obsession with food (I jokingly say I spend half my waking day thinking about my next meal) and my ability to not let the dark issues I work on to interfere with my general mood. My only evidence against the proposition is that I became a foodie sometime during my Ph.D. I think I learnt more about Carnatic music and food during those years than Physics.’

Ramana: thanks for some very interesting questions. What you‘ve said about my views on the difference between journalism and fiction is spot on. Generally speaking in River of Smoke as in Sea of Poppies, I have tried to stay within the boundaries of the possible and the credible, but since these are novels I have also allowed myself certain liberties.

My descriptions of the dishes served at the Chinese banquets in Canton are based mainly on first-hand sources – travelers’ descriptions and the like. But as to the food that merchants like Bahram ate in Canton there are no records, so I had no choice but to extrapolate.

Cantonese cuisine is one of the great culinary treasures of the world. I love it and I am sure Bahram loved it too. But it must be said that to a palate accustomed to Indian fare a regular Cantonese diet can seem somewhat bland. The foods of Sichuan and Hunan on the other hand are spicy enough for the most masaledar tastes. Bahram’s mistress Chi-mei would, I am sure, have been aware of this and she would have tried to provide him with dishes that he was sure to relish – and rather than cook them herself she would have ordered them from the countless stalls and eateries of the waterfront, among which there are sure to have been many that served Sichuanese fare.

In the Canton of the 1830s there were certainly plenty of people from Sichuan, Hunan and other parts of China. The great Chinese writer Shen Fu, who visited Canton at about the same time as Bahram, tells us in his delightful memoir Six Records of a Floating Life[i], that many cities and provinces were represented even on the flower-boats of the Pearl River:

“‘The girls on Chaochou boats dress like goddesses,’ said one of my friends. ‘We could go and see them if you like.’…Hsiu-feng said to me: ‘Across the river from Chinghai Gate there are Yangchou boats, where the girls all dress in the Soochow fashion. If we go there you are sure to find one you like.’

“Another of our friends added: ‘What the people call the Yangchou boats have only one madam, called Widow Shao; with her there is her daughter-in-law, called Big Sister. They are from Yangchou, but their girls are all from Hunan, Hupei, Kwangtung (Guangdong) and Kiangsi.’”

A place of such cosmopolitan offerings would not I am sure, have lacked for the fare of Sichuan, which is among the most popular of Chinese cuisines. Dan-dan noodles are a well known contemporary Sichuanese item: the name is probably a recent invention but I feel certain that some similar dish existed in the 19th century. From what I know of Bahram I feel certain that he would have relished these noodles (but perhaps this is only because I fully share his passion for Sichuanese food).


Leshan Buddha, Sichuan, 8th century


Chai is a more complicated matter. The name is itself of considerable significance, because the etymologies of words for tea, in many languages around the world, lead back to two places. Both are of course in China since that is where tea originated: they are Fujian and Guangdong (Canton), which are the two maritimes provinces where foreign merchants first went in search of tea and other Chinese goods.

The written character for tea is the same in all the provinces of China, but in the Fujianese dialect the pronunciation of the word is somewhat idiosyncratic (the Cantonese pronunciation is similar to the Mandarin). The English tea and the French thé, are adaptations of the Fujianese pronunciation, as are the words for tea in most European languages. Whether these languages took the word directly from the Fujianese dialect or borrowed it from an intermediary is a matter of dispute. The authors of Hobson-Jobson incline towards the view that it was taken from Malay: “Crawfurd alleges that we got this word in its various European forms from the Malay Te, the Chinese name being Chhâ… But though it may be probable that Te, like several other names of articles of trade, may have come to us through the Malay, the word is, not the less, originally Chinese, (or Tay as Medhurst writes it) being the utterance attached to the character in the Fuhkien dialect. The original pronunciation, whether direct from Fuh-kien or through the Malay, accompanied the introduction of tea to England as well as other countries of Western Europe.”

It seems unlikely to me that the English, whose trade links with Canton were so old and so extensive, would have adopted a Fujianese variant had they taken their word for tea directly from China. In my view this corroborates the Hobson-Jobson surmise of the word being borrowed from an intermediary language, perhaps Malay. I find it interesting though that many European languages have preserved slang variants of the Cantonese word – for example char in English, from which we also get charwoman and so on. For a while the two variants seem to have been battling for primacy in English. Quite possibly tea prevailed because its rival char had been widely adopted by working people. For the class that wrote dictionaries this would have counted as a mark against the word; and so ironically, the mandarins of England rejected the word favoured by their counterparts in China and opted instead for provincial variant: tea.

But the French, like the English, were relative latecomers to the China trade. The Portuguese were the pioneers and they too traded principally with Guangdong; it was in this province that they founded Macau, the first European settlement in China. This perhaps is why the Portuguese parted company with most other Europeans by adopting a Cantonese variant: chá.

In the Indian subcontinent only Sinhala, Telegu, Tamil and Malayalam preserve variants of the Fujianese original; every other language has adopted Cantonese/Mandarin derivatives such as the Bengali cha, the Hindi/Urdu chai, Assamese chah, Konkani chá etc. What does this tell us about the route by which tea – the beverage as well as the names for it – came into the Indian subcontinent? Were they imported directly from Canton via the sea routes? Or were they borrowed from an intermediary such as Portuguese, or possibly Arabic, another great seafaring language that adopted a Cantonese variant, shayy?

The latter I think is not a contender because it lacks the consonant ch (hence the sh in shayy). This pretty much rules it out as a possible parent of the Indian words which have uniformly preserved ch as the initial consonant.

Portuguese is a much more likely contender. Here is what one authoritative lexicographer, M.S.R. Dalgado, has to say[ii]:

“It is not known for certain whether tea was known in India before the Portuguese arrival there, nor to what extent the propagation of the word is to be attributed to Portuguese influence, nor by what route the other form found its way to the Coromandel coast and made its entry into Ceylon. In the old Portuguese chroniclers there are not many references either to tea or coffee. The first mention of it, according to Gonçalves Viana (Apostilas) is made by Frei Gaspar da Cruz in his Tratado da China (1569): ‘Whatsoever person or persones come to any mans house of qualitee, hee hath a custome to offer him in a fine basket one Porcelane… with a kinde of drinke which they call cha, which is somewhat bitter, red and medicinall, which they are wont to make with a certayne concoction of herbes.’ [See Da Cruz in Purchas III, 180]…

“Mandelslo, quoted in Hobson-Jobson, says in 1638: ‘In our ordinary meetings (at Surat) which we had every day, we did not take anything but Thé (tea), the use of which was very common all over India.’ But this ought to be understood in connection with the Europeans, their descendants, and some indigenous Christians; for even today, the strictly orthodox Hindus abstain from tea, and Mussulmans prefer coffee.”[iii]

“John Crawfurd alleges that the word tea in its various European forms came from the Malay Te. If it did not find its way into India through the same channel, which is little likely, Sinhalese must have received it from the Dutch thee and Tamil and Telegu from the French thé. And, in this case, it is very likely that the other Indian languages received their various forms directly or indirectly from the Portuguese chá.”

I think M.S.R. Dalgado is right: in all likelihood it was Portuguese that served to introduce the Cantonese/Mandarin word into the languages of the Indian subcontinent. The elites of the Portuguese colonies in Goa, Daman and Diu would no doubt have been the first to use and propagate the word. But I suspect that the people who brought the word into popular usage across the subcontinent were lascars – who, as always with working class people, get very little credit for their cultural and linguistic innovations.

All this being said, it still does not account for the particularities of our desi chai – which is made by boiling cheap, black tea with milk, sugar and spices. This is not ‘tea’ in the sense in which it would be understood either in China or Continental Europe: it is rather a milk-based beverage of a kind that has a long history in India – a slight flavouring of tea is the only thing that sets it apart. Who invented this drink and who introduced it to the Indian subcontinent?

Here again, I think lascars were the conduit, but of a later generation.

By the late 18th century African and Asian lascars formed a large part of the global seafaring workforce. Many of them worked on British ships, where tea was a regular item of the common seaman’s diet. Like the daily ration of grog it was served at certain hours of the day, and was much looked forward to.

This seafaring beverage was made with very cheap tea, boiled and re-boiled many times. Sugar made it palatable and milk was added when available, following the usual British practice. This drink sustained British sailors and soldiers well into the 20th century. It figures in innumerable accounts of the world wars. Not till the arrival of the tea bag did it begin to lose ground.

On most ships lascars had their own kitchens and were allotted their own rations, which included rice, daal, spices and of course tea. It is easy to imagine how the usual seaman’s char would have changed in passing through lascar kitchens, acquiring spices of various kinds, including a few pinches of garam-masala. This last was eventually to lend its name to the beverage – ‘chai-garam’.


A few months ago a historian wrote me to say that his research suggested that very few Indians other than Parsis had tasted tea before the 19th century.

In dealing with the Indian subcontinent it isn’t easy to decide on the implications of words like ‘few’ or ‘many’. In relation to the total population a number like ten million (or even twenty) could be interpreted as either. In this sense my correspondent was probably right: the number of Indians who had tasted tea in the 19th century probably were ‘few’ relative to the entire population. But if we are to speak more generally of the emergence of a practice of tea-drinking, I think it happened much earlier than he suggests.

Tea was included, as far as I remember in the rations of indentured migrants in the mid-19th century – this is hardly likely to have been the case if they had not been accustomed to the beverage.

Recently the gifted young historian Ashutosh Kumar (see my post of July 16 ‘Life on a Plantation’) sent me a list of provisions for Indian sepoys who were setting off on a foreign expedition.


Fort William 29th May 1801, pp. 713-19.

From Charles Fraser, Garrison Store keeper

To captain Isaac Humphry,

Secretary to the Military Board

[Note: theses provisions were only on ship during voyage.]

Daily ratio of No.5 provisions for the 33 regiment for each mess of five men

Monday                        2½ tls of flour 1 tls of suit 1tls of revisions 12 of now

7 ½  tls of rice


Thursday                        6 tls of pork and 1 quart of peas



Friday                                    8tls of beef 2½ tls flour 1 tls suit and 1 tls of raising each day


Weekly to each mess of 5 men

30 tls of bread 1 tls of  TEA 5tls of sugar 12 quart of vinegar 1 tls of mustard.


From this it would seem that tea was a regular item of the sepoy’s rations as early as 1801.

So to come at last to the question: Was chai garam really drunk (invented?) by Parsis in Canton in the early 19th century?

Parsis were certainly among the first to adopt the habit of tea drinking in India (many Parsi families still possess the utensils and canisters their ancestors brought back from China). But the Bombay traders who sojourned in Canton did not live there alone: they were accompanied by large numbers of servants, household staff etc (many of whom also doubled as lascar ships’ boys and so on). These men would also have been introduced to Cantonese chah. Nor did the Canton contingent represent the entirety of the subcontinental presence in southern China. There was a sizeable Goan population in Macau from the earliest years of the settlement: they too would have experimented with tea.

But as regards the invention of chai garam, the relevant passage in River of Smoke is this: ‘Bahram’s breakfast always ended with a beverage that Mesto claimed to have invented himself: the drink was made with tea leaves but it bore no resemblance to the chàh that was commonly served in Canton.’

The operative word in this is ‘claimed’. Like many great cooks, Mesto was on occasion inclined to claim too much credit for himself: in fact he had begun his career as a ship’s cook and that was how he first learnt to make chai.

But to be fair to Mesto his claim was not entirely unfounded either, for he had added an ingredient that was not to be found in any of the usual chai recipes: star anise. He does deserve some little credit for this innovation.




[i] trans. Leonard Pratt and Chiang Su-hui, Penguin Classics, 1983, 119-120.

[ii] Soares, Anthony Xavier.: Portuguese Vocables in Asiatic Languages from the Portuguese Original of M.S.R. Dalgado (translated into English with Notes, Additions and Comments), Asian Educational Services, New Delhi & Madras, 1988, p. 93.


[iii] The exact quote is: “Dans les assemblées ordinaires (à Sourat) que nous faisions tous les iours, nous ne prenions que du Thè, dont l’vsage est fort cummun par toutes les Indes.”-<-> Mandelslo, ed. Paris, 1659, p. 113.

One thought on “‘Chai’ and Dan dan noodles”
  1. A very interesting read. You take so much energy to even answer a simple query in a blog-post!! It is quite understandable the sincerity you put into visualizing and materializing the characters your create, making it all very real.. it leaves so much to imagination and yet, so less.

Leave a Reply

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