Introduction

The population figures for mandatory and Turkish Palestine are of historical interest and figure in many historical debates. The Zionist claim that Palestine was "a land without a people" is challenged by pro-Palestinian historians who cite census figures showing a substantial Palestinian-Arab population by 1914. The Zionists note that most of this increase seems to have occurred after 1880, when Jews began developing Palestine. In particular, Joan Peters ("From Time Immemorial") claimed that a large proportion of the population increase among Arabs was due to immigration. Pro-Palestinian historians try to make a case that Zionist settlement had begun displacing Palestinians before 1948.

The goal of the present is to examine the claims in the light of the best available statistical data, without supporting the contentions of either side, and without any intention either to denigrate from the tragedy of Palestinian refugees or to use the data to question Jewish claims to Palestine. The moral claims of the sides should not depend on percentages of population. In practice, I am aware that the data on this page have been used to support various partisan claims. That is precisely the sort of abuse that this material is intended to fight.  The major conclusion is "The nature of the data do not permit precise conclusions about the Arab population of Palestine in Ottoman and British times"  Anyone who pretends otherwise is deliberately misleading you. We can reach some general conclusions - Palestine was not empty when Zionists started arriving, there was some Arab immigration as well etc. But we cannot give a precise number in any case, and even if we could, it would not constitute evidence to back any moral claims.

Uncertainties in the data - Debates about the population of Palestine flourish because of the lack of good information and confusion over the meaning of census figures, and the will of partisans to distort history. Census figures of the Ottoman Empire were unreliable. Foreign residents were not counted, and illegal residents did their best to evade the census, as did people wishing to evade military services and taxes.   The population figures of the British mandate were more reliable, but there was no published census taken after 1931. Mandatory figures for the period after 1931 are based on hospital and immigration records and extrapolation, it seems. Nomadic Bedouin  were not counted or undercounted in both Ottoman and British censuses. Those who became settled in Palestine would then add to population figures. In studying the population of Palestine between 1800 and 1948, we must keep in mind that there was only one agreed-upon reliable census in all that time, which took place in 1931. The British census of 1922 was taken in less than settled conditions, and may have undercounted the population. The Ottoman figures certainly undercounted. The census data of 1922 and 1931 and the estimates based on these censuses have also been challenged but they appear to be internally consistent. That is, in the main, the number of people reported by the British mandate in 1922 and 1931 is consistent with the rates of natural increase that they reported.  The numbers given in the 1945 survey are about 100,000 or more below what would be expected based on the number of refugees and remaining population in 1948. Uncertainties in infant mortality and underreporting of births would not account for all of this discrepancy. It could be due to illegal immigration or in part to settling of nomadic Bedouins in the Palestinian Arab population.

Economics and Immigration - Under the British Mandate, which began after WWI, Jewish population increased due to immigration, especially in the 1930s. Arab population also increased at an exceptional rate. According to records, about 18,000 non-Jews  entered Palestine between 1930 and 1939 when there were more or less reliable figures. In the same period, about 5,000 non-Jews left. This does not count illegal immigration of course, or immigration prior to 1930.  Economic analyses show that by the 1930s the standard of living of Palestinian Arabs was approximately twice that of Arabs  in surrounding countries, whereas in Ottoman Turkish times it was lower than in surrounding countries. Some of the farm population may have suffered economic hardship, characteristic of any industrializing and urbanizing society, but in the main, the standard of living improved, and it improved much faster than it did in surrounding countries. There is no doubt that this improvement in conditions was an attractant for immigrants as well as resulting in improved health and larger families. Additionally, British activity in building the port of Haifa during the 1920s and in operating it during WW II undoubtedly attracted at least some immigrants. However, there is no hard evidence that more than 100,000 or 200,000 (out of about 1.3 million in all of Palestine, and about 7-800,000 in the area that was to become Israel in 1948) Palestinians had immigrated to the land that was to become Israel. It is impossible to determine at present when this immigration took place. 100,000 Arabs immigrating in 1880 would have produced many more descendants by 1948 than 100,000 Arabs immigrating in 1930.  However, since economic conditions did not improve until mandatory times, it is unlikely that the bulk of the immigration occurred under Turkish administration.

Joan Peters, in her book "From Time Immemorial," argues that most of the increase in Arab population was in fact due to illegal Arab immigration. Her figures are not accepted by most demographers and historians, including Zionists. Norman Finkelstein and others have criticized her thesis and shown evidence of poor scholarship. Finkelstein's analysis also shows that the largest increases of Palestinian Arab population occurred close to Jewish population centers in Palestine, which would argue against the Palestinian contention that the Zionists were dispossessing Arabs. We do not know if this increase was due to population shifts in Palestine or immigration from outside Palestine.  It is certain that there was at least some illegal Palestinian-Arab immigration, as noted in British mandatory reports. Immigration from Transjordan was not illegal, and was not recorded as immigration at all until 1938. Beginning in the 1920s when they built Haifa port, and especially during and just prior to  World War II, the British recruited Arab workers from the Houran in Syria and elsewhere. Arabs also came to Palestine before the war, attracted by higher wages. However, since much of the depletion of Palestinian population that had occurred in the eighteenth and nineteenth centuries was due to migration to neighboring countries, many of these returning Arabs may have been families returning to Palestine.

Refugees - The UN figure for Palestinian Arab refugees that is most often quoted by pro-Arab sources is 726,000. This number was later revised downward to 711,000 by the UN, but almost nobody pays attention to the change. On the other hand, pro-Zionist sources like to quote a much lower figure that was contained in an interim report by Ralph Bunche. The 711,000 figure may be closest to the truth, but there is no real way of knowing.

About this page - This page is the result of an ongoing analysis. It is not intended to be an exhaustive demographic study. Corrections and additions are most welcome.

IMPORTANT NOTE

Many of the figures presented on this page must be incorrect, because they conflict with other reports. Th purpose of showing these data is to examine the discrepancies. It is an abuse of the intent of this essay, and it is intellectually dishonest, to post one table or set of figures from this page in isolation, and to use those numbers to "prove" a political point about Jewish or Arab rights in Palestine.

Major Conclusions

1. The nature of the data do not permit precise conclusions about the Arab population of Palestine in Ottoman and British times, and the relative contributions of natural increase and immigration, imprecision in the counts and other issues.

2. Palestine was not an empty land when Zionist immigration began. The lowest estimates claim there were about 410,000 Arab Muslims and Christians in Palestine in 1893. A Zionist estimate claimed there were over 600,000 Arabs in Palestine. in the 1890s. At this time, the number of Jewish immigrants to Palestine was still negligible by all accounts. It is unlikely that Palestinian immigration prior to this period was due to Zionist development. Though uncertainty exists concerning the precise numbers of Arabs living in the areas that later became Israel, it is very unlikely that the claims of Joan Peters that there were less than 100,000 Arabs living there are valid.

3. Zionist settlement between 1880 and 1948 did not displace or dispossess Palestinians. Every indication is that there was net Arab immigration into Palestine in this period, and that the economic situation of Palestinian Arabs improved tremendously under the British Mandate relative to surrounding countries. By 1948, there were approximately 1.35 million Arabs and 650,000  Jews living between the Jordan and the Mediterranean, more Arabs than had ever lived in Palestine before, and more Jews than had lived there since Roman times. Analysis of population by sub-districts shows that Arab population tended to increase the most between 1931 and 1948 in the same areas where there were large proportions of Jews. Therefore, Zionist immigration did not displace Arabs. For a detailed discussion that focuses on this myth, please refer to Zionism and its Impact.

4. Historic population data in Palestine during Ottoman times and during Mandatory times show significant discrepancies. For example, figures reported in Table A-1 for 1930 population of Arabs are about 100,000 too low according to census figures for 1931

5. It is not possible to estimate illegal Arab immigration directly, but apparently there was some immigration. The total Arab immigration to Palestine recorded or estimated by the Mandate government was in the neighborhood of 45,000. Illegal immigration that was not recorded would not register in the final population figures for 1945, because those figures were estimates. We simply do not know how many Arabs and Jews there were in Palestine before the declaration of the state of Israel. It is probable that there were about 100,000 Arab immigrants into Palestine. An unknown number may also have migrated internally, from the Arab areas in the West Bank that were formerly the centers of commercial activity and population to the coastal plain and Galilee. The Arab population increase of areas with large Jewish settlement was about 10% greater than that in areas without Jewish settlement. This effect cannot be totally separated from urbanization. A population of approximately 103,000 Bedouin (1922 estimate reported in the 1927-1929 reports of the Mandatory) may have been excluded or included in different population figures as the authorities and demographers saw fit. There is no way to know how many of these Bedouin made a permanent home in Palestine or how many became part of the city population in the course of industrialization between 1922 and 1948. However,  the evidence indicates that they were in fact included in all the official population figures. This is shown by the fact that estimates of Muslim population that explicitly do not include Bedouin were significantly lower than the census figures, and by the fact that population growth is consistent with figures for natural increase if we assume that the Bedouin were included.

5. There are large discrepancies between official population figures and the  number of Palestinian refugees  -  An analysis of population by subdistricts and villages, using the admittedly incomplete data of the Palestine Remembered Web site, shows that there were about  736,000 Muslim and Christian Arabs in the part of Palestine that was to become "Green Line Israel" in 1949. There would not have been more than 620,000 refugees in 1949 if these figures are correct, since the Israeli census showed 156,000 non-Jews living in Palestine in November 1948, of whom about 14,000 were Druze. The number of refugees reported by UNRWA in 1948 was 726,000. It might indicate that an unregistered and illegal population of 100,000 was included in the refugees, or it might be due to serious and systematic undercounting of Arab population by the Mandate authorities. McCarthy suggests that there was such undercounting, yet his figures for the total population of Palestine agree with projections based on official figures for 1945.

6. There are serious discrepancies in reporting of the number of refugees. In 1949, UNRWA reported 726,000 refugees. By 1950 they reported 914,000 according to one source (McCarthy), an increase of 26% that could not come either from births or further displacement of refugees, which were negligible.

7. The city of Jerusalem has had a Jewish majority since about 1896 - The city of Jerusalem itself there was a Jewish majority since about 1896, but probably not before. The district of Jerusalem (as opposed to the city) comprised a very wide area in Ottoman and British times, in which there was a Muslim majority. This included Jericho, Bethlehem and other towns.  Within the Jerusalem district, there was a subdistrict of Jerusalem that includes many of the immediate suburbs such as Eyn Karem, Beit Zeit etc. In that subdistrict, the Jews remained a minority , with only about 52,000 out of 132,000 persons in 1931 for example.

Population of Ottoman Palestine

The population of Ottoman "Palestine" is difficult to estimate, because:

1. There was no administrative district of Palestine. Turkish census figures were for various districts, including the Jerusalem, Acco and Nablus districts for example. The Acre district included areas in Lebanon, outside the modern borders of Palestine in which there were no Jews.

2. Turkish census figures did not include Bedouins (estimated at a few thousand by Turks, but at 100,000 in the British census of 1922) and foreign subjects. A considerable proportion of the Jews retained their foreign nationality (usually Russian) in Ottoman Palestine.

3. Both Arabs and Jews avoided the Turkish census. Foreigners who were without residence permits did not want to make their presence known. Arabs and Jews wished to avoid taxes and conscription.

4. In the 19th century, only Muslims were subject to the draft, and accordingly, Muslims tended to avoid the census.

5. According to Justin McCarthy, the census tended to undercount women and children.

6. The Turkish census data were not published regularly, so only partial data are available.

As the data are ambiguous, different sources give different estimates. In particular, Zionist sources may exaggerate the number Jews in earlier years and undercount Arabs, and Arab sources  According to Bachi, (cited here)  there were  there were 489,200 Arabs (Muslims and Christians) in Palestine in 1890 and 42,900 Jews.

According to Beinin and Hajjar  the Turkish census for 1878 listed 462,465 Turkish subjects in the Jerusalem, Nablus and Acre districts: 403,795 Muslims (including Druze), 43,659 Christians and 15,011 Jews. In addition, there were at least 10,000 Jews with foreign citizenship (recent immigrants to the country), and several thousand Muslim Arab nomads (Bedouin) who were not counted as Ottoman subjects.

However, according to the data of Karpat, cited here, in  the Ottoman Turkish Census of 1893, there were 371,959 Muslims and 42,689 Christians, for a total of 414,648 Arab Palestinians, and only about 9,000 Jews. The data of Beinin and Hajar probably include subdistricts of the Acre Sanjak that are in modern Lebanon. Everyone agrees that the numbers for Jews and Muslims are far too low. Rupin (cited in the same article here) claimed there were a total of 689,275 persons in Palestine in 1893, of whom about 80,000 were Jews. This number is probably an overestimate.

 According to Justin McCarthy, in 1860, there were 411,000 Arabs in Palestine,  in 1890 there were 553,000, in 1914 there were 738,000, but in 1918 there were only 689,000. As there was no census in several of those years,  it is not clear how he draws these conclusions McCarthy tells us that these numbers have been adjusted for undercounting of women and children, accounting for the differences between McCarthy's figures and census data. The drop during the war may have been caused by famine and disease as McCarthy claims, but he doesn't note that in 1922, the British census listed only 660,641 Arab Palestinians (Christians and Arabs, see table below) nor does he explain the drop from 1918. Perhaps the earlier figures include areas of Palestine not included in the mandate or other overestimates.

By 1908, according to Dr. Hala Fattah ( http://www.jerusalemites.org/jerusalem/ottoman/1.htm ) :

   :" when Sultan Abdul-Hamid II's rule collapsed, it was estimated that the Jewish population of Palestine had risen to 80,000, three times its number in 1882, when the first entry restrictions were imposed." Other estimates put Jewish prewar population as low as 40,000 and as high as 100,000.

According to Arjan El Fassed and Lauri Irani (Originally at electronicintifada.net/historicalmyths/nosuchthing.html and no longer on the Web - 2007) in 1912 there were only 40,000 Jews and 525,000 Arabs in Palestine. However, Beinin and Hajjar claim that the "Arab population in 1914 was 683,000. By the outbreak of World War I (1914), the population of Jews in Palestine had risen to about 60,000, about 33,000 of whom were recent settlers."

The war reduced both Arab and Jewish populations to some extent, so that there were variously, according to different sources, 40,000 to about 80,000 Jews in Palestine.

Comparing some of these numbers is illuminating. The census of 1893 gives a total of 414,648 Arab Palestinians. Table A-1 below lists 469,000 Arabs for 1893, Bachi claimed there were 489,000, McCarthy estimated 553,000, and Rupin estimated about 600,000 all for approximately the same year. Likewise, as noted, there were wide discrepancies for Jews as well. Arjan Fassed and Lauri King Irani (see table below) claimed there were only 7,000 Jews in 1870, and 10,000 in 1893 (apparently taking the Jewish population figures, but not the Arab ones from the Turkish census of that year)  while Bachi estimated that there were about 42,000 Jews in 1893. Hala Fattah claimed about 80,000 Jews in 1908, while table A-1 of Arjan El Fassed and Lauri King Irani listed only 60,000 in 1914.

The data for Arab population estimates are given below. The Census of 1922 is the British Census of course, while that of 1893 is the Ottoman Census. As we can see from inspection there is no agreement between the numbers. In part this may be because they refer to different areas and some include subdistricts that were not part of Palestine after 1918. The origins of these data are not really known. McCarthy's prewar figures are probably overestimates of Arab population, even assuming great undercounting in the Turkish census.

Table 1: Comparison of different estimates of Arab Population of Ottoman Palestine

To give an idea of the variability and uncertainty in Ottoman data, Table 1-a presents estimates of population in the Qouds (Jerusalem) district, which comprised about 2/3 of the future area of Palestine

Table 1a: Ottoman population figures for the Qouds District¹

1. Data are from the statistics compiled by  Jan Lahmeyer at the Populstat   Web site using various sources.

2. The actual years are given as 1884/5, 1890/1, 1900/1, 1910/1911

3. An average value taken from two estimates by the same source.

It is very unlikely that the population increased by 44% in 5 or 6 years between 1885 and 1890, or that in 25 years the population increased 63%

The Areas of Jewish Settlement under the Ottoman empire

The principle arguments center around the population of Arab Palestinians in the areas of Palestine that were eventually included in the state of Israel in 1948. In 1948, it is estimated that these areas included some 800,000 Arabs (there is no certainty about this figure either. Joan Peters claimed that the population of these areas was about 92,000 in 1893, based on population figures for the seven Turkish subdistricts that approximately comprised Palestine in 1948. This would mean that Jews, who numbered perhaps 85,000 according to optimistic estimates, might comprise the largest single minority in that area. The origin of these figures is uncertain, since the Turkish census gave 198,000 non-Jewish persons for these sub-districts, probably an underestimate, and Cuinet, a traveler of this period, estimated about 186,000

Table 2: Arab Population of Future Area of Israel in 1893

If we assume that the initial population was 200,000 in 1893, and that there was a yearly natural increase of 2.7%, we would reach the figure of 820,000 in 1948, without assuming any immigration at all. Assuming that the Ottoman census undercounted, this is not an unlikely surmise. For all of Palestine, between 1922 and 1931, the census figures for non-Jews, correspond to an annual increase of about 2.9%, while between 1931 and 1948 they correspond to an increase of 2.0%. This difference may be due to undercounting in the 1922 census or errors in the estimate after 1931, or to drops in the actual birthrate.

If we accept Peters' figures, then we would have to assume that the shortfall was made up by an additional immigration of somewhat under 100,000 Arab Palestinians since 1893, some of whom would have had considerable offspring by 1948.  However, as Yehoshua Porath has shown (see note below) Peters' figures are very unlikely.

Population of Mandatory Palestine

A. Population Growth in Palestine

There were only two censuses taken in Mandatory Palestine, in 1922 and in 1931. All other figures for population of mandatory Palestine are based on reported births and deaths and immigration. The Anglo-American survey of 1945 provides valuable additional data for population in that year, but it too is probably incomplete. Zionists point out that data after 1931 do not reflect illegal immigration of Arabs, as well as Jews, while pro-Palestinians believe that the census omitted many Bedouin and understates the Palestinian birthrate. Justin McCarthy asserts that the census of 1922 was done carelessly, and other Palestinian sources challenge the data from 1931. Unfortunately, there is no way to "correct" the values of a census that was done carelessly and there is no reason to assume a consistent bias in one or another direction. The 1922 and 1931 censuses have arbitrary estimates of the number of Bedouin in the Negev. The numbers were not based on actual census questionnaires. Moreover, these Bedouin were not sedentary. They moved between the Sinai, the Negev and what is now Transjordan. There is no way to know what percentage of this subpopulation could be said to be permanent residents of the Negev.

Jewish population during the mandate and immigration - The census of 1922 listed 83,790 Jews in Palestine. The census of 1931 listed 174,606. The Anglo-American report of 1946 listed 608,000 Jews in Palestine. According to the Israel Statistical Abstract, there were 716,000 Jews recorded in November 1948 and 758,000 recorded at the end of the year.  It is not possible to ascertain the actual number of Jews present at the birth of the state, but the number given is generally 650,000.  Considerably numbers of immigrants entered immediately after the state was declared. A net immigration of 216,000 Jews was recorded for 1930-1939. According to the Israel Central Bureau of Statistics, about 483,000 Jews immigrated to Palestine between 1919 and 1948.

Arab population and illegal immigration - The Anglo-American report of 1945 listed about 1,222,000 Muslim and Christian Arabs in Palestine and 15,000 "others." . The mandatory blue book reports for the 1920s estimated about 25,000 illegal Arab immigrants in total that were not recorded. The 1937 Mandatory report estimated about 25,000 legal Arab immigrants over the entire period.

In addition the report notes:

"5. There has been unrecorded illegal immigration both of Jews and of Arabs in the period since the census of 1931, but no estimate of its volume will be possible until the next census is taken."

The data concerning legal immigration of Arabs were also reported occasionally in the annual reports of the Mandatory submitted between 1923 and 1938) but  in a haphazard and obscurantist fashion.

The table below summarizes approximate population growth in Mandatory Palestine

Table 3: Approximate population growth in Mandatory Palestine

1. Figures for Jewish population were estimated to include immigration. 650,000 is the accepted number. Number of others were estimated based on average rates of increase in 1922-1945. The source http://www.palestineremembered.com/Acre/Maps/Story574.html gives the number 608,250 for 1945 as a revised survey figure and this number is generally accepted. However, table A-1 and others list the survey numbers as if they are for 1946 rather than 1945.

2. These widely quoted numbers are apparently likewise based on the official estimates and were not due to a special survey. A copy of the report (abridged) that is on the Web gives only figures for 1944 (not revised for Jewish illegal immigration and of course not revised for Arab illegal immigration, which has never been estimated). It states:

Population

4. According to official estimates, the population of Palestine grew from 750,000 at the census of 1922 to 1,765,000 at the end of 1944. In this period the Jewish part of the population rose from 84,000 to 554,000, and from 13 to 31 percent of the whole. Three-fourths of this expansion of the Jewish community was accounted for by immigration. Meanwhile the Arabs, though their proportion of the total population was falling, had increased by an even greater number-the Moslems alone from 589,000 to 1,061,000.* 

Table 4: Palestine Mandate: Growth of Non-Jewish population from 1922 - 1937

The  report of the mandatory for 1937 lists population calculated according to the two census and according to estimates of population made based on immigration, emigration births and deaths in intervening years. No allowance is made for illegal immigration in these figures. These figures are given in the table below, along with calculated rates of increase for each year.