Who Owns Bogurar Doi? An Ethnography of Placemaking and Craftsmanship in Bangladesh Free

This essay is based on fieldwork conducted in Sherpur Upazilla of the Bogura district in Rajshahi Division of Bangladesh at intermittent phases in June–July 2019, as well as on a close reading of literary writings in Bangla and scholarship on caste and sweet-making. I interviewed owners and workers of doighars (shops that sell doi) in Sherpur and Bogura Sadar, observed the making of doi in designated workshops, and interviewed dairy scientists and other related professionals at the Rural Development Academy, Sherpur.

Like all craft-based occupations in South Asia, sweet-making is deeply gendered and marked by caste and religious ties. According to Achaya (2002) in Arthashastra, a curd seller was called Maithitika. Bengali literature is resplendent with the figure of doiwala (a man who sells curd) and goalini (a woman from the Goala caste group associated with milk and milk produce), clearly indicating the gendered and caste-based labor of curd making. Most of the sweetshop owners of Sherpur repeatedly told me in their interviews about the contribution of the legendary Ghetu Ghosh and his family, who were skilled in making doi. In a video on Bogurar doi, one of the sweetshop owners observes that “Ghosh-s” used to run the business, but now it is dominated by Muslims; this was a recurrent statement during the course of my fieldwork. Ghosh is one of the surnames² used by people who identify themselves as Goala—the caste group associated with the trade of milk and milk-related products. Ghosh, as I will show, continues to enjoy a mythic status of craftsmanship in sweet-making in general and doi-making in particular. It is against this backdrop that we shall revisit the question of whom is a doiwala.

In 2019, as I stepped out of one workshop, I spotted male workers attending to boiling milk, or checking the clay pots kept around the oven to set the doi. One of them lifting the chhati/matal, a half-moon-shaped wicker basket, to cover the clay pots with a mix of sweetened boiled milk with starter culture told me that I had arrived at the right time to learn about doi-making. In summer months, doi is one of the fastest-moving items in a sweet shop, and I could now get a real taste of what it took to make Bogurar doi (which requires an ample supply of quality milk and sixteen to eighteen hours of labor to prepare). When I asked the workers if they agreed that air, soil, and milk are necessary to make a good Bogurar doi, one of them quipped, “to that list add an ability to work in this really hot workshop with four ovens around which clay pots are kept for doi to set.”

Doi is the Bangla word for one of the most widely consumed fermented milk foods across the world—yogurt. It is known as mast in Iran and Afghanistan, dahi in parts of India and Pakistan, and doi in Bangladesh. At the outset, I want to clarify that in India and Bangladesh, the English word “curd” is commonly used to refer to a fermented milk product that resembles yogurt, a point emphasized across many food writings (Sanghvi 2012; Davidson 2014). In this article I feel compelled to use the terms “doi”/“yogurt”/“curd” interchangeably because many of the sources that describe this fermented milk product have used the terms “curd” and “yogurt.” For instance, the well-known food historian on Indian food Achaya (2002) uses the word “curd” to describe this food item. Achaya (2002) writes “curd” is prepared across homes from adding “leftover curd” to “boiled and cooled milk to body temperature” (57). It is an integral part of the diet of the Indian subcontinent. According to Achaya, in Vedic literature there is mention of curd served with barley in Arthashastra, and in Ain-i-Akbari there are references to the use of curd in meat marinades (58). In other words, it is more than just a dessert, especially the doi prepared from milk and starter culture. In his historical scholarship on Bengali cultures, Murshid (2005) observed that doi was consumed across the religious and class divide. Doi was a favorite food item across rich and poor households. However, the poor did not always have the luxury to consume doi. Alongside doi, ghol (a beverage made from doi, water, salt, and sugar) was also popular.

Doi is prepared in homes and sweetshops and by such multinational dairy giants as Nestle, Danone, and Epigamia, as well as homegrown brands such as Mother Dairy and Amul in India and Aarong in Bangladesh. In Bangladesh, Grameen Bank joined hands with multinational dairy giant Danone as Grameen Danone Foods Limited and has launched a fortified curd called Shokti doi that contains Vitamin A, Iron, Iodine, Zinc and Calcium. It is consumed as an accompaniment to rice, puffed rice, and beaten rice. In Bangladesh, too, doi is used to marinate meats and to prepare borhani, a doi-based beverage served with spicy food.

Bogurar doi is a form of mishti doi. Literally meaning sweetened curd/yogurt, mishti doi is indigenous to Bengal, what has come to be known today as West Bengal and Bangladesh. In the case of mishti doi, milk is boiled down and then sugar added before adding the starter culture. In India, most dairy brands, including Mother Dairy and Epigamia, have launched mishti doi along with flavored yogurts. In other words, mishti doi is an extremely popular sweet item in West Bengal and Bangladesh, and both of these geographies—particularly in Bogura in Bangladesh and in Nabadwip and Mollarchowk in West Bengal—are known for mishti doi to the degree that city names have become integral to the names of doi. What value do place names carry?

As soon as I entered the Rural Development Academy in Sherpur, one of the officials who was helping with my accommodation remarked, “It is a miracle that North Bengal has such a diverse range of sweets. Don’t you think?” The northern region of Bangladesh is known for its harsh climatic conditions—floods, river erosion, and monga/mora Karthik.³ People in the northwestern region of Bangladesh reportedly face “seasonal poverty…a near-famine situation of acute scarcity of employment during the Bangla months of Bhadra-Kartik (September–November),” which is called monga/mora Karthik (Paul, Hossain, and Ray 2013). These conditions periodically force people in need of work to migrate to Dhaka, where they face derogatory remarks around deprivation, drought, and floods. Alongside climatic challenges, many districts in northern Bengal share a border with India and were known as important trade routes along what is today the Indo-Bangladesh border. Owing to its distance from Dhaka, “the nerve centre of policy, decision-making and politics in Bangladesh,” the northern region, according to some, has faced a “policy distance” (CGIAR Research Program on Water, Land and Ecosystems [WLE] 2017). It is against this backdrop that we should locate the interlocutors’ comments on the diversity of sweets. The northern region of Bangladesh, comprising the Mymensingh, Rangpur, Sylhet, and Rajshahi divisions, is known for sweets made from milk and milk derivatives. Of the four divisions, according to a “sweets-mapping” undertaken by Sweets of Bengal—an art café located in an upscale neighborhood of Dhaka—Rajshahi Division is known for an array of sweets. Apart from Bogurar doi, which is prepared from milk, sweets in other districts of Rajshahi are made from chhana (a coagulated mass of milk after separation of whey water), kheer (thickened milk), and khoa (desiccated milk). Almost every district of Rajshahi Division is known for a sweet. Shibgunj upazilla, of the Chapainawabgunj district, is known for chomchom (an oblong-shaped, chhana-based sweet boiled in sugar syrup and coated with desiccated khoa). Kanchagolla, a pristine, white sweet made from a semi-cooked mix of chhana and sugar syrup, is synonymous with Natore. Another local specialty is a small, disc-shaped pynara made from desiccated milk and sugar mix. Non-milk-based sweets, especially mihidana laddu, a round sweet made from gram flour and sugar, is popular across sweetshops in Naogaon and Natore in Rajshahi Division. In other words, there is another side to northern Bangladesh alongside food scarcity, monga, floods, and river erosion.

Is this, the parallel coexistence of the sweets industry and drought, yet another side of the “Bangladesh paradox” (Chowdhury 2019)? According to Chowdhury (2019), “paradox” appears as a recurring “motif when discussing Bangladeshi politics, economy, and history” (7). The idea of the paradox provides us an entry point to understand why it is important to pay attention to emic discourses about the quality of milk, water, and atmosphere that make Bogura’s doi unique.

Perhaps the top priority is quality milk. “Ekhanker dudh bhalo” (Here, you get good milk) is a common phrase I heard in sweetshops as well as from dairy experts at the Rural Development Academy. Most discussions pointed to high-quality cow milk and the importance of quality cow breeds. On my further probing about quality cow breeds, one official commented, “Do you know Rabindranath Tagore was responsible for introducing the Sahiwal breed [to] Bangladesh?” Another dairy scientist observed that the introduction of the Sahiwal breed has been one of the significant contributions to doi production. I did not come across references to whether or not Nobel Laureate Rabindranath Tagore—whose songs occupy the status of national anthem in India and Bangladesh—introduced the Sahiwal breed in undivided Bengal, which came to be known as East Pakistan in 1947 and Bangladesh in 1971. Incidentally, the national anthem of Bangladesh was written after the first partition of Bengal in 1905, which was adopted when Bangladesh became an independent nation in 1971.

Tagore’s contribution to the cultural life of pre-partition Bengal does not end with his writings. This was clearly evident in my discussions with Uday Shankar Biswas, an associate professor of folklore, who insisted that I visit Patisar, a village in Atrai upazilla of the Naogaon district (one of the districts in Rajshahi Division), where Tagore had started the Patisar Krishi Bank in 1905 so that peasants could take out loans from banks rather than middlemen. Patisar was one of the zamindari estates of Tagore in what was then East Bengal, and his emphasis on self-reliance and the reconstruction of village society concerned not only economic empowerment but also education and health. Tagore initiated “experimentation with self-reliant village development” (Rahman 2006) in three of his zamindari estates: Shilaidaha, Kaligram, and Sriniketan. The most successful was Kaligram Pargana because of the able leadership of a schoolteacher, Atul Sen (2006: 232). Apart from these initiatives in his zamindari estates, according to Mehedi Hasan (2002), Tagore played an instrumental role in initiating what is today known as Bangladesh’s bathanbhumi, “a special type of built area for raising and maintaining cattle in the rural areas of the country” in Shahjadpur upazilla. The legend goes that in 1896, a Girishchandra Ghosh of Goala Samaj demanded some tax-free land from then zamindar Rabindranath Tagore. The article further observes that Rabindranath not only donated tax-free land but also provided Girishchandra Ghosh, a quality breed of cow. Today, bathanbhumi is spread over Sirajganj and Pabna (Hasan 2002), and Tagore’s mansion in Sirajganj, just two hours from Bogura, remains a tourist attraction like Patisar.

Most discussions on quality breed of cows point to the introduction of the Sahiwal breed. A scientist at Rural Development Academy explained this further: “You have to understand why your respondents are referring to Sahiwal breed—because there has been extensive cross-breeding of indigenous breeds with Sahiwal breed,” she said. When I asked her about the origin of the Sahiwal breed, she alerted me to Ahmed and Islam’s (1987) report that shows that artificial insemination probably began in 1935. In this context, one must remember that the Royal Commission of Agriculture was set up in 1928, and that Lord Linlithgow, chairperson of this commission, vehemently argued for the improvement of cow breeds. As I followed the movement of the Sahiwal breed, I realized that the British were instrumental in introducing the breed to Kenyan farmers as well. Since its introduction in Kenya, the Sahiwal breed has become so common among pastoralists that they never realized it was a “foreign” breed, observes Jitendra (2019) in a Down to Earth article. The Sahiwal breed is known to be tick- and drought-resistant and can yield up to five litres of milk per milking session.

The Sahiwal breed is named after a city in the Punjab province of Pakistan. In colonial documents, the cows are referred to as one of the best milking breeds of the Indian subcontinent. In a monograph on the Sahiwal breed, Prakash and colleagues (2005) draw on Sir Arthur Olver’s (1938) monograph on important breeds of cattle where he argued that the Sahiwal breed is closer to the cattle of Afghanistan. This breed of cattle is especially known as a dairy breed. In their monograph, the authors mention that around 200 years ago, the King of Bikaner in Rajasthan invited Sahiwal breeders from Montgomery (now Sahiwal) in Pakistan. These breeders settled in Rajasthan, but after partition they shifted back to Pakistan.

This returns us to the discussion on crossbreeding of Sahiwal cows with indigenous breeds in Bangladesh. During my interactions with experts at the Rural Development Academy, I learned that one of the perceived local breeds, the Pabna breed, was the result of crossbreeding of another local variety with Sahiwal cows. As my Academy source went on to explain various types of crossbreeding, especially with Sahiwal cows, I asked how, given the extent of crossbreeding, one made sense of when people involved in doi-making remarked, “We prefer milk from ‘desi goru.’”

Desi goru means indigenous cow breeds. My source observed that there were breeds such as the Red Chittagong Cow indigenous to Bangladesh. True to its name, the cow is red in color, and the breed is found more or less everywhere in the Chittagong district and the Chittagong Hill Tract region. She further explained that Pabna cattle are a favorite among locals. Before I could confuse Pabna cattle as yet another example of a cattle breed named after a place, she reminded me that this too was a crossbreed with Sahiwal. Another breed that has been around for some time is Sindhi cattle. In Bogura, North Bengal Grey is one of the preferred indigenous breeds and is also shown as desi, implying indigenous and local. The coat color ranges from gray to white, and skin color ranges from black to brown. Apart from these identifiable indigenous breeds, I also learned about various nondescript indigenous breeds.

Armed with this knowledge, as I tried to crosscheck the meanings of desi goru with other interlocutors, especially wage laborers in workshops dedicated to doi and some suppliers, they pointed out that they preferred desi goru (indigenous) cows because they could withstand the heat and humidity. Pointing to the sweat on my forehead, one of them commented, “Sister, if human beings cannot endure the heat with fans, air conditioning, think about the cows. When they go for grazing, they come across various things that I am not aware of. The soil quality has undergone a change. We add pesticides to the soil to grow crops. How will I know whether or not the grasslands where I take my cows are not bhejal [adulterated]?” He went on to add, “Usually a desi goru can fight back. The goru, like us, has been part of this soil, air, water and knows how to adapt—you know what I mean.” As I started engaging with literature on indigenous cattle breeds of Bangladesh, especially the Red Chittagong Cow, North Bengal Grey, and Sahiwal, I encountered an underlying common factor—all these breeds could withstand heat, were tick-resistant, and according to some cattle owners, less prone to ill health.

Requesting anonymity, one of the cattle owners remarked, “You can feel from the milk if it is from a healthy cow…have you tasted cow milk?” He gave me a cup of milk to taste in a peak summer afternoon. Initially, I was hesitant, as I was full. We had just been to a doi-making workshop where we had had a portion of doi and another milk-based dessert called ksheersha (a dish made from reduced milk). Rahamat, the rickshaw puller who had brought me, said, “Apa (Sister), drink up. Don’t worry. You will not fall ill.” I gave in and drank. As I finished the cup, the owner asked me, “Tell me if you can feel the difference. Is your mouth dry? Can you feel an aftertaste of something creamy?” I nodded. He said, “Now you know why doighars (implying shops) dedicated to preparation and sale of doi come to us. You need quality milk with a high fat percentage.” Like the worker in the sweetshop, he insisted that shops rely on desi dudh, implying milk from indigenous breeds, and once anyone has tasted desi dudh, they will never touch “packet milk.” Packet milk is a blanket term used for treated milk, implying pasteurized milk of varying quantities (500 milliliters to 1 liter) and qualities (e.g., full cream, fat-free), sold in plastic pouches in designated outlets and supermarket stores.

I was told that milk from cows indigenous to Bogura can be identified by its sweet taste, a dark cream color, and the thick layer of milk cream that settles after boiling. Milk from crossbreeds has a light cream hue, a thin layer of milk cream, and doesn’t taste sweet. Both karigars (artisans) and dairy scientists agree that the fat percentage was higher in desi dudh. One of the karigars advised me to take a quick trip to the market place where milk suppliers sold leftover milk. One of the challenges the milk suppliers voiced was that “in this line of business, it is very difficult to meet the exact demand. It depends on the mood of the cow. On good days, a desi goru can give three to four liters and on some days it might not.” Many of them acknowledged that to meet demand, they ended up selling milk from “Jersey, Holstein” (blanket terms used for crossbred cows), especially during peak season. One of them said, “Instead of adding water, if we mix it will still work in a sweet shop.” Another contradicted him, saying, “For doi, quality milk is essential.” He rambled the names of sweetshops in Sherpur and Bogura Sadar where the quality of milk was checked with a lactometer and then by the ustad (implying the head craftsman) who could ascertain the quality of milk from the layer of milk cream. He asked me to strike off the names of sweetshops fearing it might lead to unnecessary controversy. “Everyone has to survive. Doi is one of the best-selling items.”

A reading of desi goru (cow) and desi dudh (milk) allows us to move beyond a terroir-centric reading of linking food to place. The practices of rearing desi gorus are deeply implicated in the politics of indigenous breeds, crossbreeding, and the introduction of new breeds such as Holstein Freisian cattle into the milk landscape. Across the shops, there was an awareness of Jersey cows, Holstein Freisians, and discussions on whether or not shops relied on milk from these cows were met with comments such as, “Do you want us to use a lactometer?” and “Are you here to test our milk?”

The emphasis on indigenous cow breeds, to increasing use of pesticides in soil, impurities in water, and a need to train my sight and touch to ascertain quality milk indicate two possible ways of contextualizing desi dudh, in this case locally sourced cow milk. First, quality milk must be understood as an entanglement of air, water, and soil. Second, it also hints at the overarching threat of fear of soil losing a certain quality and impacting quality of milk, and consequently doi. In Dewan’s (2019) work on food adulteration in Bangladesh, her interlocutors report that the use of high-yielding variety seeds and synthetic fertilizers had sucked the earth dry, leaving the humans devoid of shakti (power). The lack of shakti in soil and therefore food and human life experienced by Dewan’s interlocutors in Southern Bangladesh mirrors the pride in “desi dudh,” milk from indigenous cow breeds. Given the popularity of desi dudh amidst the myriad movements of breeds of cows, cross-breeding stands out. The interdependence between cattle rearers of indigenous breeds and sweetshops relying on indigenous milk is an attempt to retain a certain sense of “atmosphere,” encapsulated in the constant emphasis on jol (water), batash (air), and mati (soil). How does one perceive the craftsman—in this case, the doiwala—in this scenario? In the subsequent section I turn to the gender and caste politics that shape the figure of a doiwala.

Literally, “doiwala” implies one who prepares and sells doi—a male artisan entrepreneur. To understand the figure of the doiwala, I turn to literary constructions of the doiwala in two iconic writings of Bangla literature from the nineteenth and twentieth centuries. In these writings we find references to gendered divisions of labor, the importance of family labor, and the significance of Goalas as central to doi-making.

In Kamalakanter Daptar, written in 1895, Bankimchandra Chattopadhyay introduces us to Proshonno Goalini and Kamalakanta. Kamalakanta in an opium addict and is known as an idle Brahmin. He is brought into court as a witness to identify a cow. In the court, he refuses to enlist begging as his occupation. Proshonno intervenes and asks Kamalakanta, “Haven’t you borrowed opium from anyone?” To which Kamalakanta calls out Proshonno as a “Goala’s daughter” and insists that on court records his name is enlisted as a Brahmin who accepts invitations to Brahmanbhojon, literally meaning feeding a Brahmin by other caste groups, an activity considered auspicious. This satirical tract is a spinoff on the idea of feeding that turns into an occupation. After the court agrees to enlist his profession, Kamalakanta is asked if he knows Proshonno. He says no. Perplexed by Kamalakanta’s response, Proshonno asks how he cannot recognize her despite having her milk and yogurt. Kamalakanta retorts that of course he can recognize the milk and yogurt. The trademark of Proshonno’s milk, according to Kamalakanta, is a three-to-one ratio of water and milk. And when the beverage made from yogurt (called ghol) appears light, he knows that the yogurt is from Proshonno (Bagol 2005).

Proshonno Goalini is from Goala Samaj, which was traditionally associated with milk and milk-related professions. Goala was “a common name” for “several cow-herd castes” (Bhattacharya 1896). According to Bhattacharya (1896), there were a large number of Goalas among the Hindus who were associated with agricultural life. The surnames prevalent among the Goalas were Ghosh, Pal, Barik, Babui, and Dhali. Each of these surnames could also be used by members of other castes. For instance, “Ghosh” is also used by members of the Kayastha Samaj. However, over time, especially in contemporary Bangladesh, almost every region had a legendary Ghosh synonymous with a famous sweet of the region. In the case of Bogurar doi, it was Ghetu Ghosh. Sweet-making, like most craft-based occupations, was rooted in caste-based hierarchy. In undivided India, Modak/Mayara was considered the confectioner caste, like the Halwais in north India. However, in my visits to Bangladesh, I have repeatedly found narratives around the association of “Ghoshs” with sweet-making. When I asked one of the owners about the famed Ghetu Ghosh, he quickly referred me to Ripon Dodhi Bhandar. He said, “Maybe you should visit the Ghosh para [neighborhood]—they will be able to tell you. There is also a temple.” In undivided Bengal, it was not uncommon to have designated neighborhoods based on caste groupings. Neighborhoods such as these show the lived experiences of caste, especially its divisive nature. So “Goala/Ghosh” cannot simply be reduced to a cultural category associated with milk and doi-making. This becomes clear in Kamalakantar Daptar, where Kamalakanta sarcastically remarks at one point how he as a Brahmin man can have any relation with a Goalini, pointing to the caste hierarchy (Bagol 2005). Yet what is amiss from today’s understanding of doiwala/Goala Samaj is the family-based labor that went into milk and doi production. Women from the Goala Samaj played a significant role in the making of doi and milk cream (especially in Krishnanagar, Nadia district, and West Bengal) and the managing of cattle herds. This invisibilization of labor becomes evident in the emergence of the image of the doiwala and the idealization of male craftsmanship, especially in doighars (shops) dedicated to selling doi. Yet, women play an important role in cattle-rearing, especially “feeding the cows, cleaning up cow dung, making fuel from cow dung, and bathing the cow,” and studies show that “out of 14 hours of household chores, 6 hours are spent in cow rearing” (Dey, Islam, and Sowad 2017). Despite the involvement of women in cattle-rearing and playing an important role in sourcing milk, men remain at the forefront of selling milk and doi.

In Rabindranath Tagore’s three-act play Dakghar (1912), Amal, an orphaned boy, is forced to stay indoors when he is diagnosed with fever. He waits by the window and sees a doiwala passing by. In this literary rendition, Amal is bewildered about the doiwala’s village. Doiwala tells him his village lies on the banks of a river. Amal says that upon his recovery he wants to visit the village and doiwala should teach him the tune to which he calls out, “doi, doi bhalo doi!” (curds, curds, good nice curds!) and show him how to balance the yoke that he uses to ferry stacks of yogurt in clay pots. Doiwala advises Amal to study and become a learned person (Dutta and Robinson 1992).

The two contrasting literary representations of Proshonno Goalini and Doiwala helps us to locate doiwala as a labor form rooted in caste-based occupational hierarchy and also as an idyllic pastoral representation. However, in today’s Bogura, who is a doiwala? Does this mythic imagination of the doiwala match the mushrooming of doi shops with dedicated workshops where wage laborers work around the clock to prepare doi (yogurt)?

Every narrative on Bogurar doi begins with the mention of Ghetu Ghosh, his enduring relevance, and almost everyone directed me to visit the Ghoshpara in Sherpur. Naming of neighborhoods along caste lines, or in this case surnames alluding to castes, also show the history and present status of caste in South Asian context. The name Ghoshpara, in other words, signifies a neighborhood of people who viewed themselves as Goala—a caste group identified with the trade of milk and milk product and cattle rearing. Apart from the Ghoshs, Hindus from contemporary Odisha and Marwaris owned sweetshops in colonial Bengal, along with Mayaras, the confectioner caste of Bengal. Yet, as one digs deeper, counternarratives appear. For instance, the web portal of Akboria, one of the famous hotel chains in Bogura and also known for its doi, claims that “During the 19th century, Hindu(s) were the majority of the northern region of the country and had dominance over most of businesses; as a result, the Muslim community was treated unfairly. Since the cow is a sacred animal to the Hindus, its consumption was prohibited in the whole northern region of the then Rajshahi Division. To tackle the injustice towards Muslims, ‘Akboria Grand Hotel’ started its journey and is claimed to be the first restaurant that started selling beef commercially in the northern region.”⁴ Akboria hotels are synonymous with sourcing milk from its own dairy. Most sweetshop owners and workers in today’s Bangladesh are Muslims.

This shift in ownership must be understood in the larger sociopolitical context of the partitioning of the Indian subcontinent in 1947, the creation of Pakistan and India, which led to a mass and continuous exodus, especially on the eastern side of the border. Not only was the creation of Pakistan based on religion, but its two territories (East and West) were separated by India. The uneven development between West and East Pakistan triggered the first set of grievances, but the declaration of Urdu as a national language in November 1947 was a major setback for people in East Pakistan, and over time this cultural imposition led to a series of protests by university students and the intelligentsia. Finally, in 1971, Bangladesh became an independent nation state (Sengupta 2019). Taking a cue from Bandyopadhyay (2012), in her reading of decolonization in South Asia, Sengupta (2019) argues that it would be a mistake to treat decolonization as a “event”; rather, it is a long, drawn-out process of political transition (2019: 11). This political transition, as Sengupta rightly points out, started as a negotiation between the “colonial ruling elites” and “indigenous ruling elites” (2019) and translated into diversification of occupations, especially craft-based occupations such as sweet-making.

Despite these shifts of ownership and changes, Bogurar doi is a product of repetition that underlies craftsmanship in Bogurar doi-making. Sweetshops across Bangladesh in general and Sherpur in particular have neatly arranged glass showcases and a designated place to consume dine-in sweets and snacks. Well-known establishments such as Asia Sweets and Akboria Hotel in Bogura have dedicated space for customers to sit and enjoy sweets. Most sweetshops also serve savory items either throughout the day or in designated hours. There is a designated production space for doi. My interlocutors used two words interchangeably: karkhana (implying factory) and workshop.

As I left a sweetshop, the sales manager gave instructions to Rahmat (the rickshaw puller) to take me to the workshop to see how doi is made. The time had reached nearly 2:30 p.m. when the workers were completing the task of boiling. One of the senior workers took me to the area where a worker was boiling the milk using a dabu (a ladle with a cup-shaped front). I could see that the milk had reduced, and the worker was taking a portion of boiling milk with his right hand and pouring it on the same kadai (a cooking utensil resembling a wok) from a height. One of the techniques distinct to Bogurar doi is this reduced milk. In my interview with almost thirty workers attached to sweetshops and ten workers who cater to wholesale markets, all agreed that the first step is to boil the milk to reduce it. Boiling milk, as one of the workers attached to a well-known sweetshop told me, is labor intensive. Usually, when the milk is filled to the brim, a worker must move the ladle in circles and once in a while take the dabu to pour the milk from a height. The trick is to retain the fat and not to “let the sar [milk cream] settle at the top,” I completed the sentence. To this the worker said, “Yes, right.”

Overhearing our conversation, one of the workers chuckled and commented, “Apa [sister], be honest, how many workers did you interview before?” Another worker quipped, “Learn from didi how to make sar. Opar er technique.” Opar (that side) was often used by workers to refer to both my sense of belonging and my other field site in West Bengal. During the course of previous conversations when they quizzed me as to how I knew the colloquial names of tools, I had told them of my previous experiences of fieldwork in Krishnanagar in the Nadia district of West Bengal. These conversations often became partly banter as they tried to test my knowledge. As we watched the worker continue to boil the milk, I asked if they took turns or followed an unsaid roster to take on boiling work. Initially, most workers agreed that anyone could do any task. They made comments such as “it is not very difficult.” And slowly they also pointed out that in this line of work (doi-making), it is especially important to develop a sense of approximation by observation and repetition, or what Paxson calls “synesthetic reason” (2012). Synesthetic reason comes with experience, from repeating the task. One of the workers said, “Not all milk is the same. Even after undergoing lactometer test, you need ‘experience’ to sense how much sugar is to be added.” “We don’t taste as we go along,” observed another worker. The previous worker continued, “So we need an experienced eye to know exactly the proportion of sugar that is to be added to reduced milk and mix it well.” One of the younger workers joined in, unlocking his mobile to show me a YouTube video on Bogurar doi. “Everything is here,” he concluded. I assured him that I would watch the YouTube video and continue the conversation, and then observed as leftover curd from the previous day was mixed with the reduced milk.

One of the workers asked me to come to the area where the reduced boiled milk with sugar would be added to clay pots. The clay pots (sara) are flat bottomed. One of the important tasks before adding the mix of reduced sweetened milk with the previous day’s curd and milk is to disinfect the sara. The saras are left in the oven for some time, and then neatly arranged in circles around the burner of the oven. One of the senior workers who had just finished with boiling work joined in, asking me, “Did they tell you they will add yesterday’s curd to slightly warm milk.…This is called beej [seed]…without the seed no curd will set.” I waited for the clay pots to be ready. Two young workers took the clay pots with an iron tong, removed them from the heat, and started arranging them neatly around the oven.

By then, another worker took the mix of sweetened reduced milk and “seed” and started pouring the mix from a plastic jug into the clay pots. One of the workers asked me to take a photo of the matal/chhati made from bamboo.

As the workers finished adding the milk mix to the clay pots, they showed me the bubbles that had appeared on the top. One of them quipped, “After two hours, half of the milk we have added will dry up. After all clay soaks up.…” Two workers asked us to move as they carried the jhapi/chhati/matal to cover the clay pots. This cover is added so that the heat is equally distributed across the clay pots. One of them inspected, checking for gaps between the chhati so that “no heat escapes.”

“Didn’t I tell you to add our ability to withstand in the list of things other people told you—air, water?” observed a worker. A few hours later, the workers had to lift the chhati carefully and add sweetened reduced milk, as the clay pots soaked up quite a bit of milk. This process, one of them told me, had to be repeated twice. As I crosschecked about twice or thrice, one of the workers said it was best to write that this process of adding milk would continue until the milk was set to curd. Once the doi was set, small bubbles of milk cream deposits stiffened. Sometimes black particles from the coal oven also got mixed in. “Not all factories can afford a gas burner,” someone said.

“Repetition” is key to training of hands and eye. “See, your eye is trained to see our hands move (drawing a circle) and then imitating the dabu move; similarly, we work in ‘rhythm.’” How do we understand routine and repetition, especially in the making of a microbial product? Sennett (2008) reminds us that while describing “industrial labor,” Adam Smith called routine “mindless,” yet it is in “repetition” that the helper and worker find a much-needed “rhythm” critical to craftsmanship. Thus, Sennett concludes, “the skilled craftsman has extended rhythm to the hand and eye” (2008: 175). One of the workers challenged me: even if he gave me a foolproof recipe, would I be able to create doi that resembled Bogurar doi?

Despite such experienced hands that labored, most of the Muslim workers and owners of doighars repeatedly told me that I should visit the Ghosh para, as they might know something more. Despite the transition of the sweet-making business in Bangladesh from Ghosh to Muslim ustads and owners, the constant reference to Ghosh also reveals a mythic sense of craftsmanship that circulates via oral legends. In the case of sweet-making in Bengal, Ghosh reveals the caste-based roots of the occupation that does not disappear, even if the craft has moved to a society whose social stratification is not organized around “caste.” At the same time, this shift also challenges the mythic sense of caste-based division of labor and questions the significance of “tacit” knowledge inherent to intergenerational craftsmanship. In other words, the movement of Bogurar doi-making from Ghosh to Muslim craftsmen also shows how craftsmanship can depend on the development of synesthetic reason and a knowledge about milk, temperature, and attuning the hand and eye to “routine,” challenging caste-based hierarchy and the creation of “artisanal castes.”

Bogurar doi, as most workers and owners of sweetshops told me, is known for certain techniques such as keeping the clay pots in the oven. A few kilometers away, in Bodolgachhi, in the Naogaon district, one of the doi makers told me, “What separates Bogura from Bodolgachhi is they burn the clay pots. We don’t.” The triad of soil-water-air that lends uniqueness to Bogurar doi is shaped by the movement of people, the introduction of new breeds of cows, especially cross-breeding. The meanings of Bogurar doi are shaped by “movements” that are critical to a decolonial reading of placemaking. Firstly, desi goru (indigenous) cow breeds and quality milk, critical to Bogurar doi, is produced by confluences (Banerjee-Dube 2016), thereby challenging notions of territoriality and taste-making (Leong-Salobir et al. 2016). The second movement can be traced to caste-based craftsmanship of doi-making and the mythical status of Ghoshs in sweetmaking. The third, and most important, the everyday narratives of “repetition” (Sennett 2008) that are inherent to bodily labor performed on milk to create this microbial product. A decolonial reading of placemaking allows us to examine the “emerging” movements that shape foodways and histories of craft-based food commodities with roots in sociopolitical histories of caste, religion, and nation-building. A decolonial reading of placemaking in a nutshell allows us to tell accounts of human–nonhuman entanglements in foodways.

I thank the editorial team of Gastronomica and two anonymous reviewers for their valuable comments and suggestions that helped me to finalize the article. This research would not be possible without the Research Recurring Grant (RRG) of South Asian University in 2019. I thank colleagues at the Department of Anthropology, Rajshahi University, especially Dr. Bokhtiar Ahmed, and colleagues at Rural Development Academy, Sherpur. I remain indebted to Dr. Rajat Kanti Sur for the rich discussions on caste in Bengal and literary representations of Goala Samaj. I thank Dr. Piya Srinivasan for sharing her valuable insights through the course of completing this article.