Religion Data of Census 2011: XXXIV Indian Subcontinent

Indian Religions would soon be a minority 
in the Indian region

---

We are postponing analysis of the religious demography of the Scheduled Tribes for the remaining States of North, Central, West and South India. In these States, the Scheduled Tribes are concentrated in specific districts and the Census of India has not yet released the 2011 data disaggregated up to the district level.

In this note, we look at the changes that have taken place in the religious demography of Indian Subcontinent – the region that encompasses India, Pakistan and Bangladesh.  This region has constituted the geographic and cultural India since times immemorial and also formed the political India until the Partition of 1947.

When we first carried out collation of the Census data on religion from 1881 to 1991 in the first edition of our book, Religious Demography of India, we had noticed that the changes in the relative share of different religions during this period were so sharp that if the trend continued then the Muslims and Christians together would come to form a majority of the population of the Indian Subcontinent sometime in the second half of the twenty-first century. Correspondingly, the Hindus, Sikhs, Jains, Buddhists and others, whom we collectively referred to as Indian Religionists, would be reduced to a minority in their own civilizational region.

That conclusion of ours had led to much comment and criticism. A couple of social scientists from the Madras Institute of Development Studies carried out a rather elaborate regression exercise to prove that we of the Centre for Policy Studies, being merely physicists and not trained social scientists like them, did not know simple statistics. Fortunately, the Economic and Political Weekly, which carried that curious review article, also published our response, which showed that any careful observer of the data could reach the conclusion we had arrived at without having to go through the rigmarole of statistical regressions that a majority of the social scientists in India keep indulging in without caring to comprehend the data. The review article and our response may be read at our website here.

A more honest response came from the Prime Minister’s High Level Committee for the Preparation of Report on the Social, Economic and Educational Status of the Muslims of India, commonly known as the Sachar Committee. The Committee, in its report submitted in November 2006 to the then Prime Minister, Dr. Manmohan Singh, countered our observation that Muslims and Christians were likely to become the majority by 2051 or so by raising the counter question: “how does it matter which population is the largest.”

It, of course, would not matter to Justice Sachar. Even the tragic Partition of India on religious grounds mattered little to people at the higher echelons of society. But it mattered greatly to the millions who had to pay with their life, dignity and property because a particular religious group had become a majority in some parts of what was then India. And it matters a great deal when Hindus in say West Uttar Pradesh or in parts of Bihar, Jharkhand and West Bengal or in the Goalpara region of Assam in the current India find that their towns and villages have become overwhelmingly Muslim and there is no place left for them in their ancestral lands. Or, when various tribes in parts of Arunachal Pradesh or in Gajapati and Kandhamal of Odisha find that the option of continuing with their ancestral faith is being systematically closed around them. Our review of the Sachar Committee report may be read at our website here.

In this note, we revisit our projections of the rising presence of Muslims and also Christians in the Indian Subcontinent and show that our observation that Indian Religions would be reduced to a minority towards the second half of the current century remains plausible even when we take into account the data for 2001 and 2011 that has become available since then. The trendline with the data for these decades included is somewhat flatter indicating that the fifty percent point for the Indian Religions may be reached a couple of decades later than the earlier prediction. This is mainly because the United Nations estimates for the population of Pakistan, where no Census has been conducted since 1998, and the population figures published by the Census of Bangladesh both imply unbelievably sharp lowering of the rates of growth of the two populations strongly suggesting the unreliability of those figures. But there is no better data available for the populations of Pakistan and Bangladesh.

While discussing the data for the Indian Subcontinent, we also indicate how Indian Religions have been nearly completely excluded from Pakistan and the few Hindus left in Bangladesh are being systematically expelled from there.

---

Religious profile of India, Pakistan and Bangladesh

The historic, civilizational and geographic India of the ancient times was partitioned into India and Pakistan in 1947, at the eve of Independence from the alien British rule. Pakistan further split into two units in 1971, with the eastern wing seceding to form Bangladesh. Analysing the data for the whole of the Indian region and for the entire census period from 1881 onwards, which we propose to do in this note, presents certain terminological difficulties. Before Partition, and in the censuses of that period, the whole region was referred to as India; the two partitioned units were called Indian Union and Pakistan in the relevant historical documents. The historically correct way, therefore, is to term the pre-Partition India as ‘India’ and the post-Partition Indian component as the ‘Indian Union’. This is what we did in our publication, Religious Demography of India (Chennai 2003). That terminology, however, leads to some confusion. The term ‘India’ is now commonly used for the post-Partition Indian component, which is formally named the Republic of India. Therefore, the data and analysis for the whole of India occasionally gets conflated with that of the Indian component. In this note, we use the geographical term ‘Indian Subcontinent’ for the region that formed India before Partition and the term ‘India’ for the Indian Union component of post-Partition India.

The term Indian Subcontinent is not entirely apt, because it reduces the historic, geographic, civilizational and political entity of India to a mere geographic construct. Other aspects of the Indian region have always been significant to the students of India and Indian demography, as evidenced by excerpts from Kingsley Davis—the renowned demographer and philosopher who undertook a comprehensive compilation and analysis of the available data on the demography of the newly independent Indian nation in his Population of India and Pakistan (Princeton 1951)—given in the Appendix to this note.

There is another terminological issue that needs to be clarified. In our analysis below, we use the term ‘Indian Religions’ or IR to collectively refer to religions that have originated in the Indian Subcontinent; in our book, Religious Demography of India, we used the term ‘Indian Religionists’ to mean the adherents of Indian Religions. The term includes Hindus, Sikhs, Jains, Buddhists and followers of various tribal faiths and practices. The last were sometimes termed animists in the pre-Independence censuses. In the Indian censuses since 1951, the numerous tribal faiths and practices have been counted under the category of Other Religions and Persuasions (ORP). The category also includes minor religions of non-Indian origin like Jews and Zorastrians or Parsis. But their numbers are small: in 2011, about 4 thousand Jews and 57 thousand Zorastrians have been counted in India and there are hardly any of them in Pakistan and Bangladesh. Therefore, in our analysis, we have calculated the numbers of IR by subtracting the numbers of Muslims and Christians from the total population.

The term Indian Religions is methodologically meaningful. Conversions to Islam and Christianity in Indian Subcontinent have taken and continue to take place from the group of Indian Religions. Therefore, while studying the changing religious profile of India, it is significant to look at the relative share of Indian Religions on the one hand and of Christians and Muslims on the other.

The adherents of Indian Religions (IR) are of course largely Hindu. Thus, of 1,011 million IR counted in India in 2011, 966 million are Hindus and only 44.5 million from all other Indian Religions put together.

Changing Religious Profile: 1881-1941

Population data for the period prior to the Partition and Independence is easier to handle. The Census then was conducted for the whole of Indian Subcontinent and the numbers were compiled together up to that level; and as mentioned above, all Census data for this period has been comprehensively compiled by Kingsley Davis. While analysing the data, he also went into the inadequacies of data collection and the incompatibilities across different Censuses. In Table 1 below, we use that compilation for the total population of Indian Subcontinent and its religious composition for the period 1881 to 1941. The data for the religious composition of Pakistan and Bangladesh from 1901 to 1941 is taken from the figures published by the Census of Pakistan 1961 and Census of Bangladesh 1991, respectively; these sources do not give the data for 1881 and 1891. The figures for the remaining India are obtained by subtracting the figures for Pakistan and Bangladesh from the total of Indian Subcontinent. The procedure adopted in compiling Table 1 is explained in detail in Religious Demography of India (Chennai 2003). The Census of India has now published slightly revised figures for the population of India from 1901 to 1941, but since religious composition for the revised population has not been made available and the revisions are minor, we have left our earlier figures unchanged.

TABLE 1: Religious Profile of India, Pakistan, Bangladesh and Indian Subcontinent before Partition
INDIAN SUBCONTINENT, 1881-1941
	Population in thousands				Percent share
	Total	Indian R	Muslim	Christian	IR	M	C
1881	250,155	198,424	49,953	1,778	79.32	19.97	0.71
1891	279,575	220,344	57,068	2,164	78.81	20.41	0.77
1901	283,868	218,973	62,119	2,776	77.14	21.88	0.98
1911	303,004	231,503	67,835	3,666	76.40	22.39	1.21
1921	305,727	230,224	71,005	4,497	75.30	23.23	1.47
1931	337,675	252,404	79,306	5,966	74.75	23.49	1.77
1941	388,998	287,124	94,447	7,427	73.81	24.28	1.91
MF	1.56	1.45	1.89	4.18	-5.51	4.31	1.20
INDIA, 1901-1941
1901	238,364	206,518	29,102	2,744	86.64	12.21	1.15
1911	252,068	218,252	30,269	3,547	86.58	12.01	1.41
1921	251,365	216,342	30,739	4,283	86.07	12.23	1.70
1931	278,530	237,165	35,818	5,548	85.15	12.86	1.99
1941	318,717	269,119	42,645	6,953	84.44	13.38	2.18
MF	1.34	1.30	1.47	2.53	-2.20	1.17	1.03
PAKISTAN, 1901-1941
1901	16,577	2,641	13,904	32	15.93	83.88	0.19
1911	19,381	2,898	16,364	119	14.95	84.43	0.61
1921	21,108	3,274	17,620	214	15.51	83.48	1.01
1931	23,541	4,427	18,757	357	18.81	79.68	1.52
1941	28,282	5,568	22,293	421	19.69	78.82	1.49
MF	1.71	2.11	1.60	13.16	3.76	-5.05	1.30
BANGLADESH, 1901-1941
1901	28,927	9,814	19,113	-	33.93	66.07	-
1911	31,555	10,353	21,202	-	32.81	67.19	-
1921	33,254	10,608	22,646	-	31.90	68.10	-
1931	35,604	10,812	24,731	61	30.37	69.46	0.17
1941	41,999	12,437	29,509	53	29.61	70.26	0.13
MF	1.45	1.27	1.54	-	-4.31	4.19	0.13
Note: The figures for Indian Religions are calculated by subtracting those of Muslims and Christians from the Total. The row marked MF gives the number of times the population has multiplied in the relevant period and the change in its share in percentage points.

Table 1 displays several remarkable aspects of the changes that took place in the pre-Partition period in the religious demography of the Indian Subcontinent and the three units into which it is currently divided; we discuss some of these below:

IR lost 5.5 percentage points off their share in Indian Subcontinent

In those six decades, the share of Indian Religions in the population of Indian Subcontinent declined by 5.5 percentage points, while the share of Muslims rose by 4.3 and that of Christians by 1.2 percentage points. The population of IR in this period multiplied 1.45 times while that of Muslims grew by 1.89 times.

But the IR share increased by 3.8 percentage points in Pakistan

Surprisingly, in the region that forms Pakistan now, the share of Indian Religions increased by nearly 3.8 percentage points, while that of Muslims declined by 5.0 percentage points. The growth of Christians in this part was also higher than in Indian Subcontinent as a whole and in the other two parts. Their share in the Pakistan part rose by 1.3 percentage points, compared to their gain of 1.0 percentage points in India and 0.13 percentage points in Bangladesh.

Population of Pakistan grew faster than the other two parts

Total population of Pakistan in this period grew considerably faster than that of Indian Subcontinent as a whole and of the other two parts. The population of Pakistan, however, grew by 1.71 times, while that of the parts that form India and Bangladesh now grew by 1.34 and 1.45 times, respectively. The higher growth of Pakistan was because the British, as a matter of deliberate policy, were bestowing special attention to expanding irrigation and cultivation in that part of the subcontinent to the neglect of the Ganga plains.

Rise in the share of IR in Pakistan was because of rising number of Sikh cultivators

This expansion of irrigation and cultivation attracted Sikh cultivators to the area, which led to the rise in the share of IR that we have noticed above. Census of Pakistan clubs the figures for Sikhs in the category of “others”. That category comprises mainly Sikhs. During 1901-1941, population in this category multiplied 5.63 times with the numbers rising from 3.14 lakh to 17.68 lakh. The share of “others”, therefore, rose from 1.89 to 6.25 percent. The share of Hindus, however, declined from 14.04 to 13.44 percent.

The share of IR declined both in India and Bangladesh

While the share of IR increased in Pakistan, because of the large number of Sikh settlers, their share declined in both Bangladesh and India. In the former, Indian Religions formed nearly 34 percent of the population in 1901; their share declined to less than 30 percent in 1941. In the India part of the subcontinent, the share of IR declined by 2.20 percentage points, while the Muslims gained 1.17 and Christians 1.03 points in their share.

Changing Religious Profile: 1951-2011

Getting the population numbers for Indian Subcontinent for the period after Partition and Independence is not as straightforward as for the previous period of 1881 to 1941. Since 1951, the Censuses of the three divided units have been conducted separately. And, while the Census of India has maintained regularity and commanded widespread confidence in its count, it has not been so in the other two units, especially in Pakistan. Let us, however, begin with the available Census numbers for the populations of the three units.

Table 2: Enumerated Population of India, Pakistan and Bangladesh after Partition
INDIA, 1951-2011
	Population in thousands				Percent share
	Total	Indian R	Muslim	Christian	IR	M	C
1951	361,088	314,934	37,728	8,425	87.22	10.45	2.33
1961	439,235	381,565	46,940	10,729	86.87	10.69	2.44
1971	548,160	472,516	61,418	14,225	86.20	11.20	2.60
1981	683,329	586,336	80,293	16,700	85.81	11.75	2.44
1991	846,421	720,031	106,737	19,654	85.07	12.61	2.32
2001	1,028,737	866,349	138,188	24,200	84.21	13.43	2.35
2011	1,210,855	1,010,790	172,245	27,820	83.48	14.23	2.30
PAKISTAN, 1951-1998
1951	33,703	538	32,732	433	1.60	97.12	1.28
1961	42,880	630	41,666	584	1.47	97.17	1.36
1972	62,462	1,119	60,435	908	1.79	96.75	1.45
1981	84,254	1,389	81,554	1,310	1.65	96.80	1.56
1998	132,352	2,540	127,720	2,093	1.92	96.50	1.58
BANGLADESH, 1951-2011
1951	41,933	9,599	32,227	107	22.89	76.85	0.26
1961	50,840	9,801	40,890	149	19.28	80.43	0.29
1974	71,478	10,223	61,039	216	14.30	85.40	0.30
1981	87,120	11,358	75,487	275	13.04	86.65	0.32
1991	106,315	12,088	93,881	346	11.37	88.30	0.33
2001	123,851	12,415	111,079	357	10.02	89.69	0.29
2011	144,044	13,392	130,205	447	9.30	90.39	0.31
Note: For 1972, the Census of Pakistan counts total population as 65,309 thousand persons, but gives religious composition for only 62,462 thousand. The remaining 2,847 thousand persons were mainly from FATA province. Since the total population of Pakistan in any case needs large adjustments, we ignore this omission.

Adjustments and Corrections for India

In India, the Census could not be conducted in Jammu and Kashmir in 1951 and 1991 and Assam in 1981. In addition, some smaller areas remained uncounted in some of the Censuses. In all these cases, the Census of India has published estimated populations of the uncounted parts. In arriving at the figures of India in the Table above, we have calculated the religious breakup of these uncounted populations by interpolation.

Adjustments and Corrections for Pakistan

The regular decadal Census of Pakistan was not conducted in 1971, the year when East Pakistan separated from West Pakistan to form the new country of Bangladesh. Pakistan conducted that Census in 1972 and held its regular Census in 1981. The next Census was conducted only in 1998 and the decadal Census due in 2011 has not been undertaken yet. Also the population in at least the earlier decades is widely believed to have been grossly undercounted. In view of this, and the non-availability of any Census figures for 2011, we use the estimates published by the United Nations in the World Population Prospects edition of 2015 for the total population of Pakistan in the decades following Partition and Independence. For calculating the religious break-up of the population, we take the shares of different communities as enumerated in different Censuses and listed in Table 2 above. We further assume that the percentage shares as counted in 1972 reflect the religious profile of 1971, apply the shares as counted in 1981 to the estimated population of 1991, and the shares as counted in 1998 to the estimated total populations of 2001 and 2011. These assumptions are unlikely to introduce any serious error because the relative shares of Indian Religions, Muslims and Christians have not changed much since 1951 with Muslims forming between 96 and 97 percent of the population throughout this period. In Table 3 below, we have recalculated the population of Pakistan and its religious break-up on these assumptions.

Adjustments and Corrections for Bangladesh

Up to 1971, Bangladesh formed the East Pakistan province of Pakistan. The Censuses of 1951 and 1961 were conducted along with Pakistan. The decadal Census due in 1971 was held in 1974 and the next four decadal Censuses have been held regularly. To account for the inadequacy of the earlier counts that we have noted in the case of Pakistan above, the Census of Bangladesh, in 1991, published revised estimates for the total population of the previous decades. In Table 3 below, we apply the relative shares of different communities as enumerated in the different Censuses and given in Table 2 above to this revised total population to obtain the religious break-up of the corrected populations.

Indian Subcontinent

The total population of Indian Subcontinent and its religious breakup in Table 3 below has been obtained by adding the figures for India, Pakistan and Bangladesh.

Table 3: Corrected Population of India, Pakistan, Bangladesh and Indian Subcontinent after Partition
INDIAN SUBCONTINENT, 1951-2011
	Population in thousands				Percent share
	Total	Indian R	Muslim	Christian	IR	M	C
1951	443,230	325,651	108,554	9,026	73.47	24.49	2.04
1961	540,446	392,887	136,042	11,517	72.70	25.17	2.13
1971	678,735	483,724	179,704	15,307	71.27	26.48	2.26
1981	853,933	599,388	236,306	18,239	70.19	27.67	2.14
1991	1,068,510	734,527	312,246	21,737	68.74	29.22	2.03
2001	1,293,870	881,475	385,604	26,791	68.13	29.80	2.07
2011	1,528,569	1,027,515	470,041	31,013	67.22	30.75	2.03
MF	3.45	3.16	4.33	3.44	-6.25	6.26	-0.01
INDIA, 1951-2011
1951	361,088	314,934	37,728	8,425	87.22	10.45	2.33
1961	439,235	381,565	46,940	10,729	86.87	10.69	2.44
1971	548,160	472,516	61,418	14,225	86.20	11.20	2.60
1981	683,329	586,336	80,293	16,700	85.81	11.75	2.44
1991	846,421	720,031	106,737	19,654	85.07	12.61	2.32
2001	1,028,737	866,349	1,38,188	24,200	84.21	13.43	2.35
2011	1,210,855	1,010,790	1,72,245	27,820	83.48	14.23	2.30
MF	3.35	3.21	4.57	3.30	-3.74	3.78	-0.04
PAKISTAN, 1951-2011
1951	37,976	606	36,882	488	1.60	97.12	1.28
1961	45,988	676	44,686	626	1.47	97.17	1.36
1971	59,690	1,070	57,753	868	1.79	96.75	1.45
1981	80,692	1,330	78,107	1,255	1.65	96.80	1.56
1991	110,634	1,824	1,07,089	1,721	1.65	96.80	1.56
2001	141,282	2,711	1,36,337	2,234	1.92	96.50	1.58
2011	173,670	3,333	1,67,591	2,746	1.92	96.50	1.58
MF	4.57	5.50	4.54	5.63	0.32	-0.62	0.30
BANGLADESH, 1951-2011
1951	44,166	10,110	33,943	113	22.89	76.85	0.26
1961	55,223	10,646	44,415	162	19.28	80.43	0.29
1971	70,885	10,138	60,533	214	14.30	85.40	0.30
1981	89,912	11,722	77,906	284	13.04	86.65	0.32
1991	111,455	12,672	98,420	363	11.37	88.30	0.33
2001	123,851	12,415	111,079	357	10.02	89.69	0.29
2011	144,044	13,392	130,205	447	9.30	90.39	0.31
MF	3.26	1.32	3.84	3.97	-13.59	13.54	0.06

Salient aspects of changes in the religious profile during 1951-2011

IR lose 6.25 percentage points off their share in Indian Subcontinent

As seen in Table 3, the share of Indian Religions in the population of Indian Subcontinent has declined from 73.47 percent in 1951 to 67.22 percent in 2011. Thus the trend of decline of IR that we noticed in the earlier period of 1881-1941 has not only continued but also become stronger in the later period. The corresponding gain in share, which was divided between Muslims and Christians in the earlier period, has accrued to Muslims alone during 1951-2011.

IR have lost more than 12 percentage points off their share since 1881

Indian Religions have thus lost more than 12 percentage points off their share in the population of Indian Subcontinent since the beginning of the Census in 1881. Their share of 79.32 percent in 1881 is reduced to 67.22 percent in 2011. This is indeed a very large decline for the mainstream religious group of a civilization to suffer in its own geographic region. To put the quantum of this decline in perspective, we may recall that at the time of Akbar, the proportion of Muslims in the population of the region that had come under Mughal control was no more than 16 percent, and this was after nearly four centuries of Islamic rule in many parts of India. In 1881, about three centuries later, there were less than 20 percent Muslims counted in Indian Subcontinent. That number has gone up to 31 percent in the course of 130 years of modernity. It is this sharp and continuing rise of the Muslims and decline of Indian Religions in the modern times that raises the prospect of the latter being reduced to a minority within the current century, which we discuss in the following section.

Exclusion of Indian Religions from Pakistan

Religious profile of Pakistan, 1941-1951
	T	IR	M	C	%IR	%M	%C
1941	28,282	5,568	22,293	421	19.69	78.82	1.49
1951	37,976	606	36,882	488	1.60	97.12	1.28
Population in thousands.

Another remarkable aspect of the data in the Table above is the nearly complete exclusion of Indian Religions from the area that forms Pakistan now. In 1941, there were 5.6 million IR in Pakistan forming nearly 20 percent of the total population and their share in this part had been rising since 1901. (See, Table 1). In 1951, there were only 0.6 million IR left in Pakistan. On the other hand, there was an accretion of 14.6 million to the number of Muslims. These numbers indicate the scale of elimination and forced migration of the adherents of Indian Religions from West Pakistan and the Muslims from the neighbouring parts of India that occurred as a consequence of the Partition. The number of IR in Pakistan has since increased to 3.3 million, but their share in the population remains below 2 percent. The few Hindus and Sikhs left behind in Pakistan are mere remnants of a once vibrant and growing community; they now live there more or less on sufferance with limited space in the economic, social and political life of that country. The area that forms Pakistan thus has gone out of the reach of the religious group that is the carrier of the mainstream of Indian civilisation.

Continuing expulsion of IR from Bangladesh

Religious profile of Bangladesh, 1941-1951
	T	IR	M	C	%IR	%M	%C
1941	41,999	12,437	29,509	53	29.61	70.26	0.13
1951	44,166	10,110	33,943	113	22.89	76.85	0.26
Population in thousands.

At the time of Partition, exclusion of Indian Religions from East Pakistan—the area that now forms Bangladesh—was not as complete as from what then formed West Pakistan. Even so the number of IR declined from 12.4 million in 1941 to 10.1 million in 1951, and their share in the population came down from 29.61 to 22.89 percent. Since then, their share has continued to decline rapidly from decade to decade. In 2011, IR form 9.30 percent of the population of Bangladesh; that share has declined by as much as 13.6 percentage points during 1951-2011. Thus, the exclusion of IR, which was achieved in Pakistan in one go at the time of Partition, is being accomplished in Bangladesh slowly but surely over the decades. Such expulsion of any religious group from its ancestral lands usually becomes a matter of much worldwide concern. It is indeed surprising that the persistent erosion of the share of Indian Religions in the population of Bangladesh has caused no public concern or even comment in India or in the world.

The share of Muslims in India now is above their share before Partition

Religious profile of India, 1941-1951
	T	IR	M	C	%IR	%M	%C
1941	318,717	269,119	42,645	6,953	84.44	13.38	2.18
1951	361,088	314,934	37,728	8,425	87.22	10.45	2.33
Population in thousands.

Because of the forced migration of populations between Pakistan and India at the time of Partition, the number of Muslims in the remaining India declined by nearly 5 million, from 42.6 million in 1941 to 37.7 million in 1951, and their share in the population declined from 13.38 to 10.45 percent. This decline has been more than made up in the six decades since Partition; in 2001, the share of Muslims in the population of India was already above their share in 1941 and it has risen to 14.23 percent in 2011. (See, Table 3).

Higher growth of Muslims in India and of the total population of Pakistan

These drastic changes in the religious profile of Indian Subcontinent have occurred mainly because of the higher growth of Muslims as compared to other communities and the relatively higher growth of the population of Pakistan as compared to the other two units. Between 1951 and 2011, the number of Muslims in India has multiplied 4.57 times, while Indian Religions have grown by a factor of 3.21. By sheer coincidence, the population of Pakistan has also multiplied by the same 4.57 times. The population of India and Bangladesh in this period has grown by factors of 3.35 and 3.26, respectively. The relatively higher growth of Muslims has continued since 1881. Between 1881 and 2011, the number of Muslims in Indian Subcontinent has multiplied 9.41 times while that of Indian Religions has grown by a factor of only 5.18. Christians in this period have multiplied more than 17 times, but their absolute numbers remain small.

Future Projections for the religious profile of Indian Subcontinent

Indian Religions are likely to become a minority within the current century

The data and analysis above gives us 14 data points for the share of Indian Religions and of Muslims and Christians together in the population of Indian Subcontinent. In the figure below, we have tried to obtain the best possible fit for these 14 points. The data fits a polynomial equation of third order with R²-value of nearly one. Projected into the future, the trend-lines indicate that the Indian Religions would become less than 50 percent of the population just after 2081.

In our earlier calculation in Religious Demography of India (Chennai 2003), Indian Religions were projected to reach the halfway mark 2 decades earlier. This shifting of the halfway mark has happened because the United Nations estimates for the population of Pakistan have been drastically lowered since our earlier calculation and Bangladesh has reported unbelievably lower growth of the population during the last two decades.

Lower UN estimates for the population of Pakistan

Population of Pakistan (‘000)
	WPP 1996	WPP 2015
1951	40,451	37,976
1961	51,343	45,988
1971	67,443	59,690
1981	88,197	80,692
1991	1,22,397	110,634
2001	-	141,282
2011	-	173,670
WPP: World Population Prospects

The World Population Prospects 2015 gives much lower estimates for the population of Pakistan compared to the 1996 edition. This lowering of the estimates for the earlier decades has led to lower estimates for the last two decades. The latter estimates are further lowered because, as seen in the Table below, the World Population Prospects 2015 assume a very low rate of decadal growth for 1991-2001. The sudden lowering of the decadal growth from 37.11 percent in 1981-91 to 27.70 percent in the next decade and then again to 22.92 percent in 2001-11 is difficult to explain.

Inexplicable lowering of decadal growth rates

Decadal Growth Rates (percent)
Decade	India	Pak	Bang
1951-61	21.64	21.10	25.04
1961-71	24.80	29.79	28.36
1971-81	24.66	35.19	26.84
1981-91	23.87	37.11	23.96
1991-01	21.54	27.70	11.12
2001-11	17.70	22.92	16.30
Derived from populations in Table 3.

The Census figures of Bangladesh also show a sudden and sharp decline of the decadal growth in 1991-2001 followed by a considerable rise in the next decade. Pairing down of decadal growth by more than half, from 23.96 percent in 1981-91 to 11.12 percent in 1991-2001 is hardly under­standable. And there can be no reasonable explanation for the considerable rise in growth in the next decade. The figures of India look much more reasonable with a smooth rise and decline of the growth rates. The population figures of Pakistan and Bangladesh that we have used are obviously not equally reliable. But these are the best figures that are available.

Projection for Indian Subcontinent depend crucially upon 

the growth rates of Pakistan and Bangladesh

The projection of the religious profile of Indian Subcontinent into the future crucially depends upon the rates of growth of Pakistan and Bangladesh. The considerably higher growth rate of the population of Pakistan is one of the three significant factors in lowering the share of IR in the total population of Indian Subcontinent. The other two are the higher growth of Muslims in India and the sharp decline of IR in Bangladesh. The rate of growth of Pakistan remains considerably higher than that of the Indian component. In India, rise in the share of Muslims by about 0.8 percentage points every decade and a corresponding decline in the share of Indian Religions have continued unabated. In Bangladesh, the share of Indian Religions continues to drop sharply from decade to decade. These three factors have ensured that the share of Indian Religions in the total population of Indian Subcontinent has fallen by more than a percentage point per decade on the average for the last four decades. This level of decline is higher than what was experienced in the earlier decades. These factors make our projection that Indian Religions would become a minority in Indian Subcontinent sometime in the second half of the current century highly plausible.

The likely undercounting of Christians

All these projections for Indian Subcontinent are based on the assumption that the share of Christians in the population of India is correctly counted in the Censuses of India, Pakistan and Bangladesh. As we have shown in our note on the decline of Christianity in Andhra Pradesh, there are reasons to assume that a considerable number of those who have converted to Christianity do not get counted as Christians in the Census of India. This is confirmed by the estimates of the population of the religions of the world made by Christian sources in, for example, the World Christian Encyclopaedia (New York 2001) and The World’s Religions in Figures (Chichester 2013). The former source invokes a category of secret Christians whom it names as crypto-Christians and defines them as “Secret believers, hidden Christians, usually known to the churches but not to state or secular or non-Christian religious society.” In 2000, the share of Christians and crypto-Christian together, according to this Encyclopaedia, was 5.17 percent in Indian Subcontinent and 6.15 percent in India. World Religion in Figures does not use the category of crypto-Christians, but estimates the share of Christians in 2010 to be 4.68 percent for India and 3.99 percent for Indian Subcontinent; these figures are nearly double the share of Christians counted in the Censuses of India, Pakistan and Bangladesh.

If the number and share of Christians in Indian Subcontinent is indeed as large as given in these authoritative Christians sources, then the denouement of Indian Religions becoming a minority in Indian Subcontinent obviously comes nearer.

CONCLUSION

1. The population figures for Indian subcontinent as a whole and for Pakistan and Bangladesh separately are available more reliably for the period before the Partition than the period following the Partition and Independence of Indian Subcontinent.

2. During 1881-1941, the share of Muslims in Indian Subcontinent rose by 4.3 percentage points, from nearly 20 percent in 1881 to 24.3 percent in 1941 and the share of Christians rose from 0.7 to 1.9 percent. The share of Indian Religions correspondingly declined by 5.5 percentage points from 79.3 to 73.8 percent.

3. During that period, Indian population as a whole grew rather slowly, with the population of the Indian subcontinent multiplying only 1.56 times in six decades.

4. Even so, the population of the region that came to form Pakistan later grew faster than that of India and Bangladesh. Population of Pakistan multiplied 1.71 times between 1881 and 1941, while that of India and Bangladesh grew 1.34 and 1.45 times, respectively.

5. Surprisingly, while the share of Indian Religions declined in the Indian Subcontinent, and also in India and Bangladesh, in Pakistan, it rose by 3.8 percentage points, from 15.9 percent in 1881 to 19.7 percent in 1941.

6. This rise in the share of Indian Religions in Pakistan was entirely because of the Sikh cultivators settling in the new areas that were being brought under irrigation and cultivation in this part of Indian Subcontinent to the exclusion of others. The share of Hindus in this part in fact declined from 14.04 to 13.44 percent during 1881-1941.

7. Decline in the share of Indian Religions was the most pronounced in Bangladesh. Their share in the population of that part came down from 33.9 to 29.6 percent. In the part that came to form India later, the share of Indian Religion declined by only 2.2 percentage points, from 86.6 percent in 1881 to 84.4 percent in 1941.

8. In the post-Partition period from 1951 to 2011, the Censuses of Pakistan have been very irregular and their population figures are considered to be gross undercounts. Similar problems seem to afflict the Censuses of Bangladesh, though these have been more regular than that of Pakistan.

9. In view of these difficulties, we use the figures from World Population Prospects 2015 for the total population of Pakistan and obtain its religious composition by assuming the shares of different religions to be as counted in the available Censuses. This, of course, does not fully account for the inadequacies of the Censuses of Pakistan and Bangladesh.

10. The most striking aspect of the figures of 1951 to 2011 is the nearly complete exclusion of Indian Religions from the region that what was constituted first as West Pakistan and became the whole of Pakistan later in 1971. Indian Religions formed nearly 20 percent of the population of this region in 1941; their share got reduced to less than 2 percent in 1951 and it has remained below 2 percent since then.

11. In Bangladesh part of the subcontinent, Indian Religions formed nearly 30 percent of the population in 1941. After Partition, they were not excluded or eliminated from this part as thoroughly as in Pakistan. But their share declined to less than 23 percent in 1951.

12. The adherents of Indian Religions have continued to be forced out of Bangladesh after 1951. Their share in the population has continued to decline from decade to decade and has reached 9.3 percent in 2011. Indian Religions have thus lost 13.6 percentage points off their share in the population of this part since 1951 and more than 20 percentage points since 1881.

13. There was a rise of about 3 percentage points between 1941 and 1951 in the share of Indian Religions in India as a consequence of the forced migration of populations that occurred at the time of Partition. That rise has been more than neutralized by the persistent growth in the share of Muslims in the six decades since Partition. The share of IR in India in 2011 is 83.5 compared to 84.4 percent in 1941 before Partition and 87.2 percent in 1951 following Partition. The share of Muslims has correspondingly risen from 10.45 percent in 1951 to 14.23 percent in 2011.

14. In the period following Partition and Independence, the population of all three components into which Indian Subcontinent has been divided has grown considerably faster than in the earlier period. But, as in the earlier period of 1881-1941, in the post-Partition period also, the population of Pakistan has grown faster than the other two units. Between 1951 and 2011, the population of Pakistan has multiplied 4.57 times while those of India and Bangladesh have grown by factors of 3.35 and 3.26, respectively.

15. Because of the relatively higher growth of the population of Pakistan, higher growth of Muslims in India and the continuing decline of Indian Religions in Bangladesh, the share of Indian Religions in the whole of Indian Subcontinent has declined continuously from 1881 onwards. They formed 79.3 percent of the population of Indian Subcontinent in 1881 and form 67.2 percent of the population in 2011.

16. The trend of decline indicates that Indian Religions shall be reduced to a minority in Indian Subcontinent just after 2081. This is about two decades later than what we had predicted in our book Religious Demography of India on the basis of data up to 1991.

17. This shifting of the point when Indian Religions would become a minority by about two decades has happened because the United Nations estimates of the population of Pakistan published in the 2015 edition of World Population Prospects are distinctly lower than in the 1996 edition and these estimates assume a sharply and inexplicably low decadal growth for 1991-2001 and 2001-11. The population figures given by the Census of Bangladesh also imply sharply and unreasonably lowered decadal growth in the last two decades.

18. Notwithstanding the difficulties associated with making a reasonable determination of the populations of Pakistan and Bangladesh in the post-Partition period, our assessment that Indian Religions shall become a minority in Indian Subcontinent within the current century remains highly plausible. Because, the three factors responsible for the imbalance in the growth of Indian Religions and others—higher growth of the population of Pakistan, sharp decline in the share of Indian Religions in Bangladesh and higher growth of Muslims in India—seem likely to persist for several decades.

19. The denouement of Indian Religions becoming a minority in Indian Subcontinent would arrive even earlier if the assertion of authoritative international Christian sources that the number of Christians in India and the Indian Subcontinent is nearly twice that counted by the Census authorities happens to be correct.

Appendix

Remarking on the geographical and political unity of Indian Subcontinent and its extraordinary richness, Kingsley Davis says:

“To a marked degree, in the past, the boundaries of geographical and political India have been coterminous. There are points of discrepancy that must be kept in mind, but they are few and unimportant. A more complex question is what lay within the boundaries.

“…If a man started at the extreme southern tip of India and walked in a straight line to the most northern border of Kashmir, he would cover 2,000 miles. …If he crossed the region the other way, starting at the western border of Baluchistan and going to the eastern boundary of Assam, he would cover nearly 2,200 miles… Pakistan and Indian Union together embrace approximately 1,581,000 square miles, more than half the area of United States, or four-fifths that of Europe exclusive of Russia. This made pre-partition India the most extensive unit in the British Empire, for she was exceeded only by Canada and Australia. In fact she contained almost fifteen percent of the territory of the Empire. 

“If she had been an independent nation, India would have been the ninth world power in terms of area controlled. This does not mean that the land is ninth in value. On the contrary, it is one of the world’s richest domains far more valuable than either Canada or Australia. It is probably the third most gifted of the world’s regions with respect to industrial capacity, and the second or third with reference to agricultural resources. But in sheer area alone it is big enough.”

And a little later, he describes the cultural and civilizational unity of India in the following terms:

“Indian ideas and institutions, taken as a whole, resemble those of no other people. They have a peculiar shape and flavour of their own. They have tended to transform and absorb any foreign elements that trickled into the region; for India, though politically conquered by outsiders, was never ...culturally conquered.

“This peculiar culture has to some degree penetrated and pervaded nearly every part of what is geographically India. It has everywhere been affected by local, indigenous variations… But neither the geographical nor the social barriers inside the subcontinent have been sufficient to prevent the widespread diffusion of a common, basic culture, which, despite great variation, is peculiarly Indian.”

The above quotes are from Kingsley Davis, The Population of India and Pakistan, Princeton University Press, Princeton 1951; p.8 and p.12. For more on the geographic and cultural unity and richness of India, see our publication, Timeless India Resurgent India: A celebration of the land and people of India.

Books and Publications cited

A. P. Joshi, M. D. Srinivas and J. K. Bajaj, Religious Demography of India, Centre for Policy Studies, Chennai 2003; also, Religious Demography of India: 2001 Revision, Centre for Policy Studies, Chennai 2005.

Kingsley Davis, The Population of India and Pakistan, Princeton University Press, Princeton 1951.

World Population Prospects, 1996 and 2015, United Nations, New York 1998 and 2016.

D. B. Barrett, G. T. Kurian and T. M. Johnson, World Christian Encyclopedia, 2^nd edition, 2 Volumes, Oxford University Press, New York 2001.

T. M. Johnson and B. J. Grim, The World’s Religions in Figures, Wiley-Blackwell, Chichester 2013.

J. K. Bajaj and M. D. Srinivas, Timeless India Resurgent India: A celebration of the land and people of India, Centre for Policy Studies, Chennai 2001.