You are here: Home Agenda Malnutrition The career of hunger: Critical reflections on the history of nutrition science and policy-Part 1

The career of hunger: Critical reflections on the history of nutrition science and policy-Part 1

By Veena Shatrugna

Nutrition research in 1920-30 'extracted' about 10-15 nutrients out of nearly 900 foods: carbohydrates, proteins, fats and vitamins. By 1950, Indian scientists were estimating people's requirements based on their own largely vegetarian diets, prioritising cost and recommending a diet of cereal for the nation and excluding animal protein. This exclusively cereal diet underlies the profile of malnutrition and disease today

Rapid advances in analytical techniques in the late-19th and early-20th century discovered the nutrients contained in foods such as grains, milk, pulses, vegetables, meats, fish, nuts, etc. This was the necessary first step to classifying foods as rich in carbohydrate, protein, vitamins and minerals, and in oils and fats. It was known by this time that proteins, carbohydrates and fats were the energy (calorie)-yielding foods necessary for daily activity: simple work when seated (sedentary), while standing (usually moderate), or walking and bending, or running (mostly heavy). It was also demonstrated that a person spent energy even at rest or while sleeping. This was called basal metabolic rate (BMR) essential for the heart to keep beating, breathing, maintaining body temperature, etc. Proteins were found to be necessary for laying down muscle, bone and other tissue during growth and to repair the wear-and-tear of tissue in later stages. Vitamins and minerals were discovered to be important for physiological functioning of the body. They were seen to work as lubricants that help the body function more efficiently, and as detoxifiers that remove unwanted substances, and to help speed up reactions in the body.

By 1937, over 300 foods had been analysed and categorised in what is now recognised as the early classic Health Bulletin No 23, 1st edition (hereafter called the Bulletin). Dr W R Aykroyd, the lead author of this landmark publication, was the director of the Nutrition Research Laboratories, Coonoor, under the Nutrition Research Fund Association of the Government of India. The Bulletin was produced to summarise nutrition information so that administrators, doctors and other functionaries could work with the concept of balanced diets for people. While the substantive title of the Bulletin, The Nutritive Value of Indian Foods and the Planning of Satisfactory Diets, reflects the confidence that nutrition science had in its ability to determine the kind of food (to be) consumed by a nation, the initiative as such is doubtless a product of political and intellectual ferment during that period.

The Bulletin sets out broad principles for carbohydrate, protein and fat requirements, borrowing from the work of the expert commission of the League of Nations (1936). The emphasis is on calorie requirements in relation to work. This is achieved by acknowledging the relationship between food needs of workers and the intensity of work done, and then translating this food needed to calorie requirements. It thus lays the ground for later scientific research and policy pronouncements that similarly assume the primacy of calories. The justification for expressing requirements first in terms of calories and second in terms of proteins is clarified by the authors who gave energy needs priority. Drawing from the report of the League of Nations, it recommends a 2,400-calorie allowance for a sedentary male or female, with added intake for each hour of light, moderate and heavy work (Aykroyd, 1937). It is interesting to note that the figure of 2,400 calories mentioned probably for the first time in an official document, comes to haunt research in India for the next 70 years. Cheap energy foods like cereals thus begin to find a central place in discussions on calorie requirements and in the planning of diets.

While the Bulletin sees that cereals give bulk and provide satiety, and are the staple diet of Indians, it also recognises the importance of animal proteins. It expresses the doubt that perhaps no combination of vegetable proteins is adequate for healthy development. The higher biological value of animal protein had already been argued by the League of Nations report. The Bulletin asserts that growing children and pregnant and lactating mothers need more protein. It proposes that calculations of cheap balanced diets must include adequate quantities of protein, even though it would be difficult to do so under financial constraints. This important caution was ignored in the following decades.

It is salutary that in 1937 this document does not construct a diet but points to the fact that many poor people eat ill-balanced diets because they cannot afford milk and other flesh foods. Throughout the text it advocates milk for children, replacement of milled rice by parboiled rice to maximise retention of nutritional value in the grain, and addition of oil or ghee to children's diet. While the Bulletin appears to represent, in the best faith, the cutting edge of pure nutritional science in the 1930s, it is also likely that the nascent knowledge it dealt with was too young for anyone to corrupt, misrepresent or to begin cutting corners with. It was also too early to understand the ramifications of this knowledge for governmental policy.

This publication was the most popular document of the Nutrition Research Laboratories and was reprinted many times, revised every time new scientific information was generated (the 5th edition came out in 1954).

It was in 1944 that the first officially recommended diet was formulated based on the principles laid down in 1937, in the report of the Subcommittee on Nutritional Requirements headed by Aykroyd (Aykroyd, 1944; hereafter called the 1944 report), set up by the Nutrition Advisory Committee of the Nutrition Research Fund Association. This committee was constituted in response to World War II, when it became clear that there were no institutions in place to address problems of food deficits caused by failure of imports, war demands, drought, cyclones, transport difficulties, crop disease, etc. The food department of the Government of India had been set up only in 1942 to deal with the urgent task of food procurement and distribution. The above committee was to draft a food policy that was based on the science of nutrition.

The 1944 report uses evidence produced by the League of Nations Health Organisation and prescribes calories needed by men for different intensities of work, but stresses essential nutrition through the non-cereal portion of the diet. The importance of proteins for growth in children, for mothers during pregnancy, and for the replacement of tissue wear-and-tear in adults is recognised. There is also a note of caution that proteins derived from vegetable foods had less value for the body than proteins derived from animal foods. This is in contrast to the centrality cereals achieve in official proposals and nutritional truths by the '70s and '80s when 80% of calories are estimated as coming from cereals. In recent times, both industry and nutritionists have endorsed using cereal as a dietary vehicle to be fortified with synthetic vitamins instead of introducing non-cereal nutrient-rich foods like meat, eggs, milk, etc.

Biological value of protein

The most important issue raised by the Bulletin and reiterated in the report of 1944 was regarding what is known in nutritional science as the biological value (BV) of proteins, "…the more closely the amino acid make-up of a protein resembles that of the
(human) tissues the greater its value," (Aykroyd, 1937:5). In addition, "…more important than the total protein content of the diet is the proportion of protein of high biological value. In general, protein derived from vegetable foods is of less value to the body than protein derived from animal food," (ibid, 4-5). Eggs and milk protein were set as the standard with a BV of over 90. The BV of pulses was around 60 and that of cereals falls between 60 and 70. The 1944 report also reiterated the importance of protein from animal sources, but assumed that the staple food of Indians was rice in the south and wheat in the north, and constructed a diet. Since energy is the major component of food required for the body it was obvious that this could be derived largely from grains, pulses and some other foods.

In order to ensure adequate protein intake, the group tried to include milk, fish, meat and eggs, in addition to vegetables and fruit. The crucial element in this report is an insistence on 10 oz (280 ml) of milk, hoping to double the recommendation to 20 oz as soon as possible.

This recommended dietary allowance (RDA) as it is now known is commendable coming as it did during the scarcity and hunger of the war periods.

Consumption units for national food requirements

The 1944 report also had a mandate to suggest policies for food production in India. They constructed another table of an estimated annual national requirement of cereals, pulses, milk, fish, etc (in million tonnes) using a coefficient to express the quantity of food consumed by women or children as a fraction of men's consumption. This is what is popularly known as the consumption unit (CU). Recognising the problems with this method the report adds, "…in case of certain protective foods such as milk, children require more and not a fraction of the adult requirement," (1944 report:2). There were other problems too, for example the recommended intake of a miner's wife would be greater than that of a clerk's wife, even when both did the same kind of work, because the husband's food consumption was greater due to an occupation demanding heavy manual labour (Cathcart and Murray, 1931). This anomaly led to much confusion in research done over the next 20 years. In recognition of these problems the CU was used to plan food and specially calorie requirements for populations and not at individual levels. It was also not considered suitable for proteins or other nutrients.

Implications of the Aykroyd legacy

Nutritional research under Aykroyd, though brilliantly innovative and deeply committed to the welfare of Indians, opened a Pandora's Box: concepts like consumption units, biological value of proteins, RDA based on calories, calorie needs of workers, the importance of animal proteins, etc, which became subjects for scientific research and nutritional policy. Over a short period these concepts were recast and deployed in administrative initiatives that, we will argue, systematically transformed the diets of the poor in India to cereals devoid of nutrients.

The consequences of this nutritional depletion have been far-reaching, and are responsible for a large measure of (our present) ill-health. We look at two such implications in the concept of calculated diets, and the search for milk substitutes.

Calculated diets

One of the most ambivalent results of Aykroyd's nutritional research was the idea of calculated diets. While it could be argued that calculation of the food requirements of a nation is necessary to determine volumes of agricultural production, it provided another handle for scientific policy and administration: it was now possible to look at food, the measurement of its composition and the calculation of its norms, as scientific facts divorced from the daily lives of people. Scientists and policymakers now had unlimited power to create artificial diets through simple calculation and unleash them on an unsuspecting population in the name of science, cost-effectiveness and the national interest. Since RDA computations were based largely on calories, cheaper foods rich in calories could be put together to provide 2,400 calories or 1 Man Value. For example, it was quickly understood that since almost all foods such as cereals, eggs, sugar, mangoes, pulses, oils, etc contained energy, scientists could theoretically advocate cheap but adequate sources of calories with either 650 gm of cereals (or 650 gm of sugar) and negligible amounts of pulses, vegetables or oil, thus satisfying the 2,400 calories norm at least cost.

Search for milk substitutes -- vegetarian sources of proteins

With heightened interest in low-cost vegetable sources of proteins that were nutritionally as useful as milk proteins, the biological values of a large number of foods were calculated. In the late-1930s, '40s and '50s, experiments were carried out on children from economically backward communities where one group was given milk powder (1 oz providing 10 gm of protein) and the control groups received an extra helping of the usual rice and vegetables. It was found that children given milk had significant increases in height and weight when compared to the control group (Aykroyd and Krishnan 1937; Someshwar Rao, 1961). A daily addition of 1 oz of skimmed milk powder proved to be a valuable supplement to cereal-based diets. Aykroyd and Krishnan (1937) too had investigated the value of cheaper foods like cooked soya beans, and found it inferior to skimmed milk. This failure of soya was ascribed to the low biological value of soya proteins, and all studies on soya were terminated. Meanwhile, with advances in technology, processed milk from various oilseeds, pulses and vegetable products was also tried.

In 1955, a special report on milk substitutes of vegetable origin was published by the Indian Council of Medical Research (ICMR, 1955). This report reviews all the studies using milk made from various vegetarian products such as soya, groundnut, pulses, from mixed plant products, and concludes that such processed milk should only be used in regions where it has been found difficult to rear mulch cattle, as in Assam and the Western Ghats. They however do not comment on the additional cost of such processing.

By 1955 it was clear that whatever the sources of vegetable proteins, when they were compared to milk protein in trials feeding sick or undernourished children, it was obvious that milk protein was far superior.

These conclusive findings notwithstanding, the search for cheap, and preferably vegetarian, solutions to the country's state of undernutrition resulted in a paper titled 'Treatment of Nutritional Oedema Syndrome (Kwashiorkor) With Vegetable Protein'
(Venkatachalam et al, 1956). This paper argued that though skim milk protein may be unsurpassed in its biological value in treatment of Kwashiorkor, underdeveloped countries could not use it as a basis for large-scale solutions. It conducted an experiment comparing skim milk protein and other vegetarian substitutes. The results of this study clearly pointed to the superior efficacy of the skim milk protein when evaluated by the two biochemical criteria adopted for assessing satisfactory recovery: the rise in serum proteins and albumin. In spite of this, the authors concluded, "the 'slight inferiority' of vegetable proteins should not obscure the fact that remarkable clinical improvement almost as striking as with skim milk was noticeable in cases treated with these diets". They further state that, "the real importance of this study would lie in that it has revealed such satisfactory therapeutic possibilities with cheap vegetable protein diet in this disease," (Venkatachalam et al, 1956:544).

Though the authors' conclusions were at variance with the findings in the empirical data of the paper, the scientific community remained silent; after all, the nation's economic burden came before charity, and the poor ought to be satisfied with a cheaper vegetarian solution to a protein deprivation that made them ill in the first place.

This was the first major attempt to wean the nutritionists away from advocating milk for poor, sick undernourished children, and justifying inferior quality foods, thus shrinking and redefining the range of foods for the children of the poor in India.

The economics and culture of nutritional research in free India

The RDA debates

The debates on the sources of proteins and their biological values were still raging when a revised RDA was presented in 1958 (Patwardhan, 1960). This report does not recommend a diet but presents evidence for fine-tuning the requirements of calories and proteins for Indians in relation to weights, heights, age, type of work, gender, etc. Patwardhan draws on the FAO report (1957) which suggests that since no single figure of calorie requirement can be applicable to all individuals, it would be advisable to define the calorie requirements of a reference adult first, and then apply corrections for age, weight, sex, occupation, etc. The FAO report (1957) defines a reference man as one who weighs 65 kg doing a moderately heavy job for eight hours, sleeps for eight hours with another eight hours spent on recreation, self-care, walking around, socialising, etc. A diet of 3,200 calories is recommended for this reference man, with a proviso that it is balanced (it is implicit that calories are derived from many food sources). The normal woman is defined as weighing 55 kg, working at home, or at an occupation for eight hours, sleeping for eight hours, and spending the rest of her time on recreation, walking, sitting, or doing domestic chores. Her requirement is fixed at 2,300 calories, again with the proviso that it comes from a balanced diet. Patwardhan uncritically reproduces the 'standard reference man', a model of the post-Industrial Revolution worker, and then attempts to apply 'corrections' for India, arriving at a figure of 2,770 calories for men and 1,950 calories for women. It is surprising that scientists did not see that the completely different work patterns in (largely rural) India would not be addressed by corrections and factors. It would need original research leading to a fresh statement of the problem.

In the case of protein requirements, new evidence had come in that though vegetable proteins had low biological value, when two or three vegetarian foods are consumed in one meal the biological value of the combination can become as good as milk protein. (A normative value of 100 is given to milk or egg protein. Individual vegetable proteins have a BV of 50 to 70, but a rice pulse diet or a rice curd meal in the ratio of 2:1 or 3:1 [cereal protein: pulse protein] can improve the BV of food to even 80 or 85.) The report asserts, "People do not consume cereals or pulses alone to meet their requirements of protein, and hence there is little chance of deficiency of a particular essential amino acid setting in through a diet, provided the total protein intake is adequate. Judicious combination of proteins from different sources will reduce the deficiency of the limiting amino acid which is otherwise present from a single source, etc,"(Patwardhan, 1960:25). A little thought makes it obvious that the "people" referred to in the report belong to a particular class and caste; not many poor people were consuming rice and pulse proteins in the ratio of 2:1 in 1960.

Without cutting down on calorie requirements, Patwardhan reduced protein requirements by about 32%, ie, from 82 gm to 55 gm for men and 67 gm to 45 gm for women, expressing the confidence that all people get their proteins from many sources, and thus implicitly putting an end to the debates on the BV of proteins. Once the ghost of low BV of proteins from vegetarian foods was laid to rest, a revised and updated publication in 1968 confidently states: "In devising cheap well-balanced diets in India, economic considerations often preclude the inclusion of milk or other animal foods in adequate amounts… A judicious mixture of vegetable foods like cereals and pulses can be cheap and at the same time can provide nearly as good an amino acid pattern as that of the costly animal food," (Gopalan, 1968:6-7).

This balanced diet was separated for the vegetarian and non-vegetarian groups and the latter were given an allowance of 45 gm of pulses and 100 gm of milk with added 30 gm of meat or fish and 30 gm of eggs.

Another committee to revise the recommended dietary allowances was set up in 1978 (ICMR, 1980). The preface states: "The most important change has been with regard to suggested balanced diets… (which) have been formulated using linear programming techniques to arrive at the least-cost formulations." This diet is based on cereals as the major source of calories and proteins, with reduced milk and no separate diet provided for non-vegetarians. It does not recommend fruits, flesh foods, eggs, nuts and oilseeds… in the name of economy. In hindsight, it is difficult not to be critical about this eradication of flesh foods from normative diets, in spite of their proven superior efficacy as proteins for the body. The entire debate on RDA is directed at finding the theoretically adequate but most economical solution to a difficult problem. However, the specific governmental solution arrived at clearly draws on a culture of vegetarianism common to the planners who thought on behalf of the nation. Food options and crucial dietary diversity for the majority of poor, dalits, BC, tribals, minorities, etc, in fact for 80-85% of the population, had been closed in the structure of plan thinking. Thus, it was enough to provide for distribution of cereals with little scientific consideration about what these cereals were eaten with.

An expert committee was constituted in 1988 to revise the RDA, and special attention was paid to the requirements of energy, fats and other trace elements (ICMR, 1990). Occupations were classified into various intensities and energy requirements for different kinds of work provided.

Over the years, there was stiff opposition from trade unions, institutions such as hostels, and other locations where the ICMR reports with RDA were used for planning diets based on the least-cost recommendations of earlier reports. The workers, for example, did not want the scientists to exert a downward pressure on wages by suggesting a cheap balanced diet devoid of flesh foods, fruits, nuts, eggs, etc. In response to this opposition, a balanced diet was not provided with the RDA document in 1988. However, other publications of the ICMR keep the least-cost balanced diet alive, for purposes of research and dissemination (NNMB, 2006; Gopalan et al, 1989) (1). The consumption unit was used for calculation of adequacy in all large-scale surveys carried out by the NNMB.

Reading the RDA debates and the changing recommendations over the decades it becomes obvious that when adults were advised to consume over half-a-kilo of cereals to meet their calorie needs, young children in the age-group 1-3 years and 3-5 years were required to consume a quarter-kilo of cereals to satisfy their calorie needs. The CU of children is 0.4 and 0.5, when compared to the norm of 1 for an adult male. The amount of milk recommended for children was not more than 300 ml, and as in the case of adults no recommendations were made for eggs, fish, flesh foods, oilseeds and fruits. The children were expected to eat an adult diet in quantities proportionate to their age. There was minimal allowance of milk (proteins) for their growth, though the actual consumption of milk in all the surveys is less than 100 ml. If children have managed to grow to their present heights, it is clear that they do so despite the RDA that the government has declared is adequate for them.

'The myth of protein gap'

The work on the biological value of proteins from vegetarian sources had, as we have seen, put its full weight behind the consumption of vegetable proteins as against milk proteins or even animal proteins, by advocating a ratio of cereal-to-pulse intakes. As a result, the story of 'poor quality' of cereal and pulse protein ends and RDA for proteins is drastically reduced (Patwardhan, 1960). Vegetarianism for the poor had been endorsed scientifically; but the next step takes the country hurtling down a cereal trap.

In the late-'60s, with food shortages and famine-like conditions in India there was pressure from the West that India accept food aid and specially milk powder proteins for starving children. In response to this, the Nutrition Research Laboratories published a review of diet surveys carried out on children subsisting largely on cereals and reported that 92% of the children surveyed had calorie deficits while only 35% of them were on an inadequate level of protein intake when compared to the requirements based on actual body weights (Gopalan, 1970). Since the 35% of children who were protein deficient were also calorie deficient, while the remaining 57% were only calorie deficient, it was inferred that the major bottleneck in the diets of pre-school children as far as India was concerned was calorie deficiency (which by then meant, of course, cereal deficiency!).

The paper concluded that: "These studies would indicate that if the children consume the same type of diets on which they have been subsisting, in amounts sufficient to satisfy their calorie needs, the problem of protein calorie malnutrition would be greatly minimised. It is clear by the same token that provisions of protein concentrates in the face of existing calorie deficiency would perhaps be a wasteful approach towards solving this problem in these children," (Gopalan, 1970:36). However, the paper also hints at the possible negative consequences of his recommendation when it states that: "Supplements of Vitamin A and iron are, therefore, essential," (ibid). The argument continues: "There has been, unfortunately, an overwhelming emphasis on protein concentrates and special protein formulations in the prevention and control of undernutrition… In the light of recent reassessments of protein and calorie requirements and in view of 'precise data' now available with regard to the actual diets of poor segments of pre-school child populations in India, it is clear that the major deficiency in the diets of pre-school children in India is calorie deficiency. Under these circumstances, the obvious approach would be to promote increased production and facilitate better distribution and utilisation of existing conventional foods rather than a diversion of such foods for the preparation of sophisticated food formulations," (Gopalan, 1970:37).

Paper calculations thus allowed the conclusion that children consuming more of the usual diets would get adequate calories and proteins. The low-cost scientific recommendation for children was a largely cereal-based diet with inadequate or virtually no pulses, milk, eggs or meat. The calculations were based on 6-8% protein in every 100 gm of cereal. If a child could eat 300-400 gm, or better still half-a-kilo, of cereal, he would actually get the necessary 25-30 gm of protein. Clearly, this diet was not alert to the fact that children are unable to eat large amounts of cereal. Their stomachs are too small for that bulk. They need high-calorie foods (1 gm cooked cereals provides 0.5 to 1 calorie, whereas 1 gm of fat provides 9 calories). The giants of nutrition research who computed the most economic options could not see that at least 30-40% of children's calorie requirement must be derived from fat.


Nutrition research has always prided itself on the scientific basis of its theories and the objective, rational logic behind its proposals. Yet, this section demonstrates development priorities were never far from scientific thought. There is no other explanation why the scientifically derived nutritious diet should be the cheapest. There are specific mutual effects of this close stitching of government and nutritional science.

Firstly, research is always applied, governmental and policy-oriented, carrying the hazards of such an orientation: cultural bias, governmental imperatives, international honour, and lack of critical distance. Secondly, nutritional research begins to speak both with the might of governmental authority and the majesty of scientific rationality. Thirdly, governmental policy receives the backing of science in support of its priorities, and therefore ultimately for the specific details through which it articulates its developmental goals: thus, the poor need only eat cereals.

(Veena Shatrugna joined the National Institute of Nutrition (NIN), Hyderabad, as a medical graduate, and spent 34 years there, retiring as Deputy Director. Here, she was witness to the big debates raging on nutrition such as the Myth of Protein Gap, the Green Revolution and the National Nutrition Programmes. She has worked and published in the areas of nutrition in pregnancy, lactation, women's work, energy requirements, osteoporosis, and child nutrition and growth. She helped form Anveshi, a centre for women's studies in Hyderabad)

This article is excerpted from 'The Career of Hunger: Critical Reflections on the History of Nutrition Science and Policy', by Veena Shatrugna, in Towards a Critical Medical Practice: Reflections on the Dilemmas of Medical Culture Today, edited by Anand Zachariah, R Srivatsan and Susie Tharu on behalf of the CMC-Anveshi Collective, Hyderabad: Orient Blackswan; 2010. Reprinted with permission of the author and editors


1 Gopalan, C et al. The Nutritive Value of Indian Foods was published in 1971, revised in 1989, and reprinted 10 times from 1991 to 2007

Infochange News & Features, July 2012