By Eric Vandenbroeck and co-workers
The Making Of The Modern Middle East
Part Five
To know the context of what
follows start with the
overview here, and for reference list of personalities involved here.
As we have seen the Sykes-Picot
negotiations of 1916, had agreed to cede most of greater Ottoman Syria to the
French zone of influence, although only the coastal area (i.e., today’s
Lebanon) was supposed to be under direct French rule, with the inland portions
under 'independent' Arab administration, which in practice meant Faisal
and crowned ‘King’ Hussein’s other sons. Theoretically, there was to be a kind
of border between the two zones stretching along a line drawn through Damascus,
Homs, Hama, and Aleppo, each allotted to the Arab side— though these cities
were still squarely in the French 'zone of influence.' But how was France to
exert predominance over areas now occupied by British troops? Complicating
these questions further was the Woodrow Wilson factor. Because of the possibly
decisive contribution of American troops to the collapse of German morale on
the Western front, along with the financial leverage U.S. banking institutions
now enjoyed vis-à-vis the Allies indebted to them, the American president was
believed to be nearly all-powerful on the eve of the peace talks that
would open in Paris in January.
Sykes-Picot is often
accused of having divided up the Arab world, but as we shall see Mark Sykes may
have believed that his actions had the best interests of the Arabs at heart. He
believed that, if properly encouraged, it would be possible to reawaken among
the Arabs memories of a vanished greatness and bring them closer to the
community of nations.
While the carnage at
Gallipoli mounted day by day, Sir Mark Sykes was dispatched by the War Office
to visit British commanders, diplomats and imperial officials throughout the
eastern theatre of war.
In India
Lord Hardinge, opined that ‘Sykes did not seem to be able to grasp the
fact that there are parts of Turkey unfit for representative institutions.’
During his long
return sea journey from India Sykes turned his ever-wandering attention to
Iraq, concerning which he composed a lengthy memorandum on the political and
military situation. However, in the second part of that memorandum entitled
‘Indian Muslims and the War’, his thoughts returned to the subject which had
long been the main preoccupation of both himself and his chief, Kitchener – the
ever-present danger of jihad. It was fear of militant Islam which had
underpinned his belief that Britain should cultivate those elements of the
religion he construed as ‘moderate’ and susceptible of being won over to the
Allied side; and now, having witnessed signs of anti-British nationalism among
the Muslims of India during his recent visit, he merged his visceral dislike of
‘westernized orientals’ with a conceptualization of the two
main tendencies which he believed he had detected in contemporary Islam.
On the one hand,
there were the intellectual nationalists, devious, half-educated manipulators,
who were seeking to mobilize the ignorant Muslim masses against
Britain and her Allies; and on the other hand, there were the traditionalist,
‘clerical’ and ‘conservative’ forces whose sincerely held religious concepts
were not incompatible with, nor necessarily hostile to, the romantic Tory
imperialism he himself espoused. These conservative Muslims were precisely the
sort of men who might be trusted to lead the ‘friendly native states’ which he
and Kitchener were advocating; and in the person of the Sharif of Mecca he
believed they had found such a promising figure. As for those scheming
intellectual Muslim nationalists, Sykes believed they were very much like the
leaders of the Turkish CUP. Their objective was to engross all political power
in the hands of a clique of journalists, pleaders, and functionaries, to oust
the clerical element, but to retain its power to excite an ignorant mob to
massacre or rebellion when necessary … An ‘intellectual’ with an imitation
European training, with envy of the European surging in his heart … sees in
Islam a political engine whereby immense masses of men can be moved to riot and
disorder … The Muslim ‘intellectual’ uses the clothes of Europe and has lost
his belief in his creed, but the hatred of Christendom and lust for the
domination of Islam as a supreme political power remains
After leaving India,
Sykes’s first stopover was Basra where he arrived on 19 September 1915. Sykes
was informed that Captain Arnold Wilson, responsible for the Basra vilayet,
would be pleased to meet him. The meeting was not a happy one. By now, the recently
promoted Wilson had returned to full-time political duties and was living in a
cramped office at Ashar, the old Turkish customs post on the banks of the
Shatt al-‘Arab where the Ashar creek meets the Shatt and leads up to
the old city of Basra. Although he was by now quite ill, suffering
intermittently from malaria and a form of beriberi, his appetite for work
remained undiminished. ‘AT’, as he was now commonly known, had recently
acquired a great enthusiasm for paperwork, taking great pride in multiplying
files, assembling card indexes and firing off telegrams at every opportunity.
Sykes found him in full cry, dashing through an enormous pile of waiting for
papers and disposing of them one after another like a threshing machine. Sykes
could be tactless when he was expounding one of his many enthusiasms or
prejudices and on this occasion, he made it abundantly clear to Wilson that in
India he had acquired a dim view of that country’s administration and he took
an equally dim view of the government of India’s predominance in Iraq. There
was no understanding, Sykes insisted, that Iraq was an imperial concern, not
just an Indian one, and therefore the views of London and Cairo must always be
taken into account when deciding military and political policy in this
particular theater. Moreover, Sykes couldn’t understand why so little effort
was being made to win the Iraqi Arabs round to actively supporting Britain.
Surely the Civil Administration could be more active in the propaganda line,
leaflets in Arabic, that sort of thing? And couldn’t they make greater efforts
to win over local sheiks, raise guerrilla bands to attack the Turkish flanks
and so on? In spite of his position of authority in the Civil Administration,
Wilson was still only a relatively junior officer and he must have felt
constrained to suffer this tactless onslaught from his aristocratic and
distinguished official visitor. But he was later to comment with concealed
bitterness that Sykes was ‘too short a time in Mesopotamia to gather more than
fragmentary impressions’, and that ‘he had come with his mind made up and he
set himself to discover the facts in favor of his preconceived
notions, rather than to survey the local situation with an impartial eye.’ In
particular, Sykes seemed overly concerned about doing ‘justice for Arab
ambitions and satisfy France’. Arab ambitions and French satisfaction: the two
concepts seemed hardly compatible; that was precisely what was beginning to
trouble Sykes as he traveled back from India. Over the last few months, he had
begun to appreciate that the French also had ‘desiderata’ in the Middle East.
Indeed, according to intelligence, he was receiving it was clear that they had
expectations of planting the tricolor at the eastern end of the Mediterranean
to accompany their colonies in North Africa. But since Britain’s interests
would be best served by a ‘devolved’ Ottoman Empire of ‘friendly native
states’, these two national objectives were clearly contradictory. Perhaps
there would, after all, have to be some kind of agreement on ‘zones of control’
with France.
On 17 November 1915
Sykes arrived in Cairo, the next leg of his journey home. Here he was shown
some important correspondence between the British high commissioner of
Egypt, Sir Henry McMahon, and the Sharif of Mecca in which the former, on
behalf of the British government, appeared to be offering some kind of
independent Arab state to the latter if the Sharif and his four sons launched a
revolt in the Hejaz against the Turkish government. In spite of continuing
disagreements about the exact boundaries of this new Arab state – and Hussein
was angling for a kingdom of vast proportions – all the signs pointed to an
eventual revolt by the Sharif and his sons.
Thus the die had now
been cast and Britain would have to try to patch up an agreement with the
French which somehow or other satisfied both countries while at the same time
leaving Hussein with something for which he and his Arab movement would still
be willing to fight. There was no question about it: it was going to be very
difficult. Then, out of the blue, the first hint of a solution emerged – if not
a solution at least a step in the right direction. An Iraqi Arab deserter from
the Ottoman army at Gallipoli, a certain Lieutenant Muhammad Sharif
al- Faruqi, was brought in to see him.
The Arab Question And The ‘Shocking Document’ That
Shaped The Middle East
The first meeting of
the British interdepartmental committee headed by Sir Arthur Nicolson with
François Georges-Picot had taken place on 23 November 1915. The French
representative was not convinced of the importance of inducing the Emir of
Mecca and the al-Fatat Arab nationalists to side with the Entente. Austen
Chamberlain reported to Lord Hardinge that Picot had ‘expressed
complete incredulity as to the projected Arab kingdom, said that the Sheikh had
no big Arab chiefs with him, that the Arabs were incapable of combining, and
that the whole scheme was visionary.' The secretary of state for India was very
pleased. It seemed that the French delegate ‘knows his Arab well. I expect he
has sized up the Sheikh’s scheme pretty accurately. I doubt if it has any
element of solidity or that any promise will have weight with the Arabs until
they are absolutely convinced that we are winning.’¹
Moreover, French
demands – which according to Picot the French were obliged to make as ‘no
French government would stand for a day which made any surrender of French
claims in Syria’ – were rather excessive. Picot informed the Nicolson committee
that France claimed the:
Possession
(nominally, a protectorate) of land starting from where the Taurus Mts approach
the sea in Cilicia, following the Taurus Mountains and the mountains further
East, so as to include Diabekr, Mosul and Kerbela, and then returning to
Deir Zor on the Euphrates and from there southwards along the desert
border, finishing eventually at the Egyptian frontier.
Picot, however, added
that he was prepared ‘to propose to the French government to throw Mosul into
the Arab pool if we did so in the case of Bagdad’. In amplification,
Nicolson minuted that Picot had:
Intimated his
readiness to proceed to Paris to explain personally our view – and the Arab desiderata.
M. Cambon told me that he had objected to this visit, on the ground that he
would not be well received at the Quai d’Orsay were he to carry with him such
unpalatable proposals as he had suggested. M. Picot would, therefore,
communicate with Quai d’Orsay in writing. We must, therefore, await the reply.²
The French reply had
not yet been received when a telegram from Sir Henry McMahon arrived on 30
November. In this telegram the high commissioner gave his considered opinion on
Hussein’s letter of 5 November, and at the same time took the opportunity to defend
himself against Chamberlain’s charges. He observed that the Emir’s letter was:
Satisfactory as
showing a desire for mutual understanding on reasonable lines. It also affords
an opportunity of meeting the wishes of the Government of India with regard to
Mesopotamia by some change of formula, but I cannot personally think of any
formula on that subject more favorable to Indian interests than the one
employed in my former letter, without raising Arab suspicions.
With regard to
nonentity of Shereef […] Everything would tend to prove that he is of
sufficient commanding importance, by position descent and personality, to be
the only possible central rallying point for Arab cause, and sufficiently anti-
Turkish to be in great personal danger at Turkish hands.
McMahon did not fail
to point out that his negotiations with Hussein were in a quandary thanks to
the policy ‘of awaiting in Egypt the threatened Turco–German advance’. It
jeopardized ‘any attempt to secure Arab cooperation’, and made it ‘appear
unwise urging Arabs into premature activity which through want of our support
and fear of Turkish retaliation might hasten their abandonment of our cause’.
At the same time it rendered ‘alienation of Arab assistance from Turks a matter
of great importance, and we must make every effort to enlist the sympathy and
assistance, even though passive, of Arab people’. In view of this difficult
situation, McMahon proposed to reply to Hussein along the following lines:
1. Acknowledge
his exclusion of Adana and Mersina from Arab sphere. […]
2. Agree
that with the exception of tract around Marash and Aintab, vilayets
of Beirut Aleppo are inhabited by Arabs but in these vilayets as elsewhere in
Syria our ally France has considerable interests, to safeguard with some special
arrangements will be necessary and as this is a matter for the French
government we cannot say more now than assure the Shereef of our earnest wish
that satisfactory settlement may be arrived at.
3. With
regard to the vilayets of Basra and Baghdad some such arrangement as he
suggests would provide suitable solution, i.e. that these vilayets which have
been taken by us from the Turks by force of arms should remain under British
administration until such time as a satisfactory mutual arrangement can be
made.
4. Assurance
of the Shereef that Great Britain has no intention to conclude peace in terms
of which freedom of Arabs from Turkish domination does not form essential
condition. (On some assurance of this nature sole hope of successful understanding
depends).
5. Appreciation
of Shereef’s desire for caution and disclaim wish to urge him to hasty
action jeopardising Arab projects but in the meantime he must spare
no effort to attach Arab peoples to our cause and prevent them assisting the enemy,
as it is of the success of these efforts and on active measures which the Arabs
may hereafter take in our cause when the time comes that permanency of present
arrangement must depend.
McMahon concluded by
expressing the hope that the Foreign Office would be able to reply ‘without
undue delay’, but, as George Clerk minuted, the Foreign Office could not
answer until they had received ‘the views of the Government of India […] an Alexandretta
expedition has been finally decided, one way or the other, [and] having
prepared a reply […] get the concurrence of the French government’. It was
‘therefore of little use discussing Sir H. McMahon’s views now’.
The India Office
reacted first. Sir Arthur Hirtzel observed that the India Office:
Agree with Sir H.
McMahon that for the success of these negotiations some display of force is
necessary to which the Arabs can rally.
Whether such is
possible, and, if so, where and how, are questions for the British and French
governments and their military advisers.
If it is not
possible, we doubt whether there is any real use in pursuing these
negotiations. But if it is considered expedient for the sake of appearances to
do so, they should be as vague as possible regarding future commitments.
Apart from a few
minor modifications, Hirtzel approved the proposed reply, even the
suggestion to ‘disclaim wish to urge him to hasty action’. The India Office
also qualified Chamberlain’s earlier proviso that McMahon’s promises only held
good if the Arabs acted at once (see Chapter 3, section ‘Four Towns and Two
Vilayets’). This no longer applied in case ‘there is to be no display of force.
But, if there is, Arab assistance must be immediate and universal.’³
A French reply was
not forthcoming. On 10 December, Nicolson decided to wait no longer. If the
Foreign Office kept on waiting:
We shall lose much
valuable time – and it is essential to send a reply to the Shereef as soon as
possible. In regard to Syria, McMahon can say that as the interests of others
are involved he must consider the point carefully. I think a further communication
[…] will be sent later – he can then proceed to reply on all the other points.
Would you draw a telegram embodying I.O.’s views and the viceroy’s wishes – and
we should get I.O. concurrence and Lord Crewe’s the sooner we can get of this
telegram the better.
After it had been
approved by Chamberlain and Crewe, a telegram was sent to Cairo the same day:
Importance of display
of British or Allied force round which Arabs can rally is fully recognized
here, but you will realize that present situation at Gallipoli and Salonica
makes it out of the question for the moment to embark on any other expedition.
Attitude of French
government in regard to Syria is also very difficult and we have little hope of
obtaining from them any assurance that will really satisfy Arabs.
On the other hand, we
must try to keep the negotiations with the Sherif in being, and you
are authorized to reply to him as follows:
- Points 1 and 2, as
you propose.
As regards point 3,
you should say that as the interests of others are involved, the point requires
careful consideration by His Majesty’s Government and a further communication
in regard to it will be sent later.
Point 4. We should
prefer to say that His Majesty’s Government are, as the Sherif knows,
disposed to give a guarantee to assist and protect the proposed Arab Kingdom as
far as may be within their power, but their interests demand, as the Sherif has recognised,
a friendly administration in the Vilayet of Bagdad and the safeguarding of
these interests call for much fuller and more detailed consideration of the
future of Mesopotamia than the present situation and the urgency of the
negotiations permit.
Point 5. The first
[…] assurance you propose. Point 6 […] As you suggest.⁴
In anticipation of
the Foreign Office telegram, McMahon wrote a private letter
to Hardinge on 4 December in which he tried to justify his actions
with regard to the negotiations with Hussein. He claimed that the viceroy took
‘the idea of a future strong united independent Arab State […] too seriously’,
as ‘the conditions of Arabia do not and will not for a very long time to come,
lend themselves to such a thing’. Sir Henry moreover did ‘not for one moment go
to the length of imagining that the present negotiations will go far to shape
the future form of Arabia or to either establish our rights or to bind our
hands in that country. The situation and its elements are much too nebulous for
that.’ His only objective had been ‘to tempt the Arab people into the right
path, detach them from the enemy and bring them on our side’. As far as Britain
was concerned, this was ‘at present largely a matter of words and to succeed we
must use persuasive terms and abstain from academic haggling over conditions –
whether about Baghdad or elsewhere’.⁵
McMahon also sought
the support of the sirdar. Wingate was honored with a letter for the first
time. McMahon excused his negligence in answering Wingate’s letters by
explaining that he was ‘a poor correspondent at the best of times’, and that a
correspondence also was not really necessary as Clayton kept them both fully
informed of each other’s ideas and views. After this apology, he proceeded to
complain about ‘the curious and, to me, mistaken attitude which India is taking
in the matter’, as well as ‘the unreasonable and uncompromising attitude of
France in regard not only to Syria but an indefinitely large hinterland in
which she will not recognize Arab interests’. Indian and French opposition,
combined with Britain’s ‘failure to hold out a hand to the Arabs by putting a
force into Cilicia’, made it likely that Britain would ‘lose all chance of Arab
cooperation and sympathy and drive them into the enemies hands against us’.⁶
McMahon was familiar
with the French position because Alfred Parker had forwarded a report on
Picot’s meeting with the Nicolson committee to Clayton. The latter had
circulated this report, with a covering note, to Maxwell, McMahon and Wingate.
In this note, Clayton observed that the result of the meeting was ‘only what
might have been expected with M. Picot as the representative of the French
government’, considering that Picot was ‘well known as being extreme in his
ideas, and completely saturated with the vision of a great French possession in
the Eastern Mediterranean’. Clayton took the opportunity to emphasize why he
was in favor of negotiations with the Arabs. These were important, not because
they might result in the Arabs actively supporting the Entente in the war
against the Ottoman Empire – which after the dismissal of the Alexandretta
scheme was out of the question anyway – but because they might prevent the
Arabs from joining the Turks and Germans. If the latter happened, then the call
for the jihad would become effective. The great gain resulting from a
successful conclusion of these negotiations was that Britain secured the
passive support of the Arabs:
In considering the
Arab movement, too much attention has been given to its possible offensive
value, and it has to some extent been forgotten that the chief advantage to be
gained is a defensive one, in that we should secure on their part a hostile
attitude towards the Turks, even though it might be only passively hostile, and
rob our enemies of the incalculable moral and material assistance which they
would gain were they to succeed in uniting against the Allies the Arab races
and, through them, Islam.⁷
McMahon incorporated
Clayton’s note into a telegram on the Arab question that was sent to London
three days later. He informed Grey that ‘selection of Picot as their
representative on recent committee on this question is discouraging indication
of French attitude’. The French delegate was ‘a notorious fanatic on Syrian
question and quite incapable of assisting any mutual settlement on reasonable
common sense grounds which present situation requires’. As far as the
negotiations with the Arabs were concerned, ‘conditions of Arabia never
justified expectation of active or organised assistance such as some
people think is object of our proposed mutual understanding. What we want is
material advantage of even passive Arab sympathy and assistance on our side
instead of their active cooperation with enemy.’ Clerk quite agreed with
McMahon’s opinion of Picot.
The latter had ‘been
particularly chosen, for his very fanaticism’. All in all, things could no
longer go on in this fashion:
The question is so
serious that I think it must be treated between government and government, and
no longer between M. Picot and this department. This is a matter for
consideration by the War Committee and I would venture to urge that that body
should hear the views of Sir Mark Sykes, who is not only highly qualified to
speak from the point of view of our interests, but who understands the French
position in Syria today – and in a sense sympathizes with it – better probably
than anyone.
Nicolson and Crewe
concurred in this suggestion. Two days later, Prime Minister Asquith informed
the Foreign Office that ‘Sir M. Sykes might be invited the next meeting of the
War Committee. The India Office shall also be represented.’ ⁸
Enter Sir Mark Sykes
Wingate and Clayton
regarded Lieut.-Colonel Sir Mark Sykes, Bart., MP as their champion in the
London battle for an active, pro-Arab policy. On 9 December, Wingate wrote to
Sir John Maxwell that Sykes, ‘should be a powerful ally in regard to Arab
policy’, while the next day, in a letter to Clayton, he expressed the hope that
‘Mark Sykes’s arrival in London on the 8th will mean that a definite Near
Eastern Policy will be adopted without more hovering’. The Sudan agent for his
part believed that now ‘Lord K. is at home again and also Sykes […] things may
have gone better recently’.⁹
Sir Mark’s
involvement with the Middle East dated from 1890, when he, at eleven years old,
had accompanied his father on a journey through Palestine, Syria and Lebanon.
This was the first of five prolonged travels in which he ranged the Fertile
Crescent. Inspired by his travels, Sykes had written two books – Through Five
Turkish Provinces and Dar-ul-Islam – which had established his reputation as an
expert on the Middle East, even though his knowledge of Arabic was limited
seeing that he could neither read nor write the language. At the end of 1904,
Sykes had been appointed honorary attaché at the Constantinople embassy. He had
occupied this post up to the end of 1906. Most of his stay had been taken up
with another bout of traveling through the Middle East, but he had also
developed intimate relations with Gerald Fitzmaurice, the chief dragoman,
Aubrey Herbert, George Lloyd and Lancelot Oliphant.¹⁰
It had been Oliphant
who had introduced Sykes to Oswald Fitzgerald, early in September 1914. On that
occasion, Sir Mark had offered his services.¹¹ This offer had not been accepted
straight away, and for the time being he had been forced to stay with his
territorial battalion at Newcastle. In a letter to his wife Edith, Sykes had
given voice to his disappointment ‘not to be where I could be most useful,
i.e., in the Mediterranean. Is it not ridiculous the haphazard way we do
things!’ ¹² However, Sykes had finally been ordered to come to London in March
1915, and was ‘appointed at the personal request of Lord Kitchener as a member
of the Committee formed to ascertain British desiderata in Asiatic Turkey’.¹³
Besides Sykes, this
committee consisted of representatives from the Foreign Office, the India
Office, the Admiralty and the Board of Trade. It was chaired by Sir Maurice De
Bunsen, until the outbreak of war, British ambassador at Vienna. During 13
meetings, from 12 April to 28 May 1915, the commission busied itself with
deter- mining British desiderata with respect to the future of the Asiatic part
of the Ottoman Empire. These deliberations resulted in a voluminous report,
which was presented to the Cabinet on 30 June.
In its ‘preliminary
considerations’ the committee stated that ‘our Empire is wide enough already,
and our task is to consolidate the possessions we already have, to make firm
and lasting the position we already hold, and to pass on to those who come after
an inheritance that stands four-square to the world’. Against this background,
the committee opted for a scheme in which, ‘subject to certain necessary
territorial exceptions’ – Basra, Smyrna and the Asiatic part of Constantinople
would have to be ceded, respectively, to Britain, Greece and Russia – the
independence of the Ottoman Empire was maintained, ‘but the form of government
to be modified by decentralisation on federal lines’, while Arab
chiefs would be granted ‘complete administrative autonomy’ under Turkish
sovereignty.¹⁴
Sykes was unable to
append his signature to the report, because he left England at the beginning of
June. The War Office had instructed him to discuss the committee’s findings
with the British authorities in the Near and Middle East, and at the same time
to study the situation on the spot. He successively visited Athens, Gallipoli,
Sofia, Cairo, Aden and again Cairo. Sir Mark subsequently sailed for India.
There he gained but a poor opinion of the capacities of the Indian authorities,
and was angered by their attitude towards the Muslims. It seemed that the only
thing they could think of was not upsetting ‘religious susceptibilities, a
phrase which is beginning to get on my nerves’.¹⁵ Sykes’s visit nevertheless
passed off rather smoothly. His subsequent visit to Mesopotamia was not without
incidents. Nine months later, Lloyd explained to Clayton that Sykes seemed ‘to
have been amazingly tactless, and not only to have rather blustered everyone
but also to have decried openly everything Indian, in a manner which was bound
to cause some resentment’.¹⁶ Arnold T. Wilson, at the time assistant political
officer, Force ‘D’ , observed in his memoirs:
He was too short a
time in Mesopotamia to gather more than fragmentary impressions. He had come
with his mind made up, and he set himself to discover facts in favor of his
preconceived notions, rather than to survey the local situation with an
impartial eye. Whatever we were doing to change the Turkish regime, or to
better the lot of the Armenian, Jew and Sabaean minorities, had his cordial
approval – for the rest, we must do justice to Arab ambitions and satisfy
France!¹⁷
Shortly after the
receipt of Hussein’s third letter, Sykes was back in Cairo. During his third
stay at the Egyptian capital within six months, Sykes dispatched a number of
telegrams to General Callwell. To a large extent these telegrams echoed
Cairo’s point of view with regard to the Arab question: the matter was urgent
and a decision had to be taken as soon as possible; a sympathetic attitude by
the Arabs towards the Entente was of the utmost importance, if only to prevent
the dreaded jihad; a settlement of the conflicting French and Arab claims was
feasible, as was a formula protecting Indian interests in Basra and Baghdad;
and, finally, the Arabs would not act before a landing at Alexandretta had
taken place.
Enter Sherif Al-Faruqi
Faruqi knew very
well that only a tiny proportion of Ottoman army officers belonged to
al-‘Ahd: his figure of 90 per cent was pure fabrication. He also knew that his
claim that al-‘Ahd included a ‘part of the Kurdish officers’ was
misleading, to say the least – there were perhaps no more than a handful of
members who were of Kurdish origin. Faruqi certainly did not unite
the al-Fatat and al- ‘Ahd movements: that was achieved by a senior
Iraqi officer, Yasin al-Hashimi. Al-‘Ahd had never carried out propaganda
among the Arab troops: on the contrary, its members had tried as much as
possible to conceal their activities. There had been no approach to
al-‘Ahd by the Turks or Germans offering an alliance,
as Faruqi in- formed Shuqayr (the Germans had never even
heard of al-‘Ahd). And Faruqi had not been authorized by al-‘Ahd, the
Sharif or any other part of the ‘Arab movement’ to‘receive’ the British
response to their demands.
Furthermore, as
regards the ‘information’ which Sykes obtained from Faruqi during
their interview, there was no ‘Arab Committee’ in Cairo. Neither the
(non-existent) committee nor Faruqi himself was in communication with
Sharif Husayn – in fact, it was to be a further month before Faruqi
contacted Husayn and informed him of his existence. And with respect to French
influence, although Hussein was later to offer some flexibility over French
interests in the coastal region of Syria at Britain’s request, at this point in
time both he and the majority of al-‘Ahd members were strongly opposed to
any French involvement in a new Arab state, in spite of
what Faruqi may have said to Sykes.
So, comforted by the
apparent ‘reasonableness’ of the Arab movement, as relayed to him
by Faruqi, as see underneath, Sykes returned to England where, almost
immediately, he was thrust into negotiations with M. Charles François
Georges-Picot, French counsellor in London and former French consul general in
Beirut, to try to harmonize Anglo-French interests in ‘Turkey-in-Asia’. For
nine months the French had been intermittently raising this question with
Britain. So during the first week of January 1916, Sykes and Picot hammered out
a draft agreement. Finally, as a result of an exchange of letters between Sir
Edward Grey, the French foreign minister, Paul Cambon, and Serge Sazonov,
the Russian minister of foreign affairs, a secret agreement was reached among
the three Great Powers defining their respective claims on Turkey’s Asian
provinces. Its terms were embodied in a letter from Grey to Cambon dated 16 May
1916, and in due course, it was to become known as the ‘Sykes–Picot Agreement’.
On 20 November, after
an interview with Faruqi, Sykes telegraphed to London, that he:
Anticipating French
difficulty, discussed the situation with him with that in view. Following is
best I could get, but seems to me to meet the situation both with regard to
France and Great Britain. Arabs would agree to accept as approximate northern
frontier Alexandretta-Aintab-Birijik- Urfa-Midiat-Zakho-Rowanduz. Arabs would
agree to convention with France granting her monopoly of all concessionary
enterprise in Syria and Palestine, Syria being defined as bounded by Euphrates
as far south as Deir Zor, and from there to Deraa and along Hedjaz Railway
to Maan.
Sykes also informed
the DMO that Faruqi insisted that the whole scheme depended on
‘Entente landing troops at a point between Mersina and Alexandretta,
and making good Amanus Pass or Cilician gates. He further stipulated
that Shereef should not take action until this had been done.’ Sykes added that
he agreed with Faruqi. It was ‘out of the question […] to call on Shereef
or Arabs to take action until we had made above mentioned passes secure’.¹⁸ The
day before, Sykes had sent off another tele- gram in which he had suggested
possible solutions to the territorial aspects of the Arab question. A far as
the vi- layets of Baghdad and Basra were concerned, these were
‘incapable of self-government and a new and weak state could not administer
them owing to Shiah and Sunni dissension. We might agree with Arabs
to administer these provinces on their behalf allocating certain revenues to
their exchequer […] (this corresponding to their demand for subsidy).’ At the
end of this telegram, Sykes had explained that he made his suggestions because
he believed that:
The situation is
critical. I feel that Arab nationalism as such presents no danger for India now
or in future unless we confine ourselves to the canal defensive and let Turk
and German masses assemble in Syria and northern Mesopotamia and reestablish
their prestige and so work a real Jehad with Arab support.¹⁹
Small wonder Wingate
and Clayton looked forward with confidence to Sir Mark’s return to London.
Sykes did not let them down, witness the statement on the Arab question he made
to the War Committee on 16 December. After an exposition in which he stressed that
the Arab nationalists were averse to revolutionary ideologies, tolerant of
other religions and favourably disposed towards Great Britain, he
observed that, with respect to the Arab question:
If I may say so, the
chief difficulty seems to me to be the French difficulty, and the root of that,
I think, to speak frankly, lies in Franco-Levantine finance. Vitali represents
the French group which used to be at Constantinople, who is in touch with M. Hugenin,
who is a Swiss, and he is in touch with the Bagdad railway, and they have a
great many relations with Javid. They have obtained the Syrian railways, and
that very big loan of 1914, which gave them immense concessions all over
Turkey. Now that party, I feel, is working through two agencies, and is
checking the Entente policy in the Near East. One is the French cleric which is
sentiment.
When Asquith interjected
‘What is that?’, Sykes added in clarification that he was referring to the
French nationalist party:
Which is sentimental,
bearing in mind the crusades. I think that that financial group works upon a
perfectly honest sentiment. On the other side, they work on the fears of the
French colonial party of an Arab Khalifate, which will have a common language with
the Arabs in Tunis, Algeria, and Morocco […] I think at the back of all this,
the influence that is moving them, is sinister.
Sykes considered the
French financiers ‘a very evil force working two honest forces, which are
unconscious of the real purport of it’. He proposed that Britain should pursue
a policy consisting of three steps. First:
We ought to settle
with France as soon as possible, and get a definite understanding about Syria.
Secondly, to organise a powerful army in Egypt which is capable of
taking the offensive; and, thirdly, to coordinate our Eastern operations. Get that
as one machine, and one definite problem: link up Aden, Mesopotamia – the whole
of that as one definite problem for the duration of the war. If we had that I
think it is worth backing the Arabs, no matter what ground we may have lost to
the north of Haifa.
Asked by Asquith how
he would come to terms with the French, Sykes stated that: I think
that we have those two assets. I think we can play on the French colonial if we
work it well: get into the French colonial’s head what a Committee of Union and
Progress Sherif means, and point out what they have done in India and
what they might do elsewhere. I think the French clerical is quite capable of
being influenced by reason of the danger to his one asset in Syria, and if you
rob the occult French financial force of its two agencies, then, I think you
are on the high road to a settlement.
In answer to a
question by Lloyd George, Sykes repeated his opinion that the Arab question
should first have to be settled with France before any military action could be
contemplated. With respect to that, he observed that Egyptian military opinion
‘strongly [held] the idea of making a landing at Alexandretta’, which was
confirmed by Kitchener.²⁰
In the course of the
subsequent discussion, Asquith wondered what military value attached to the
Arabs. Echoing Clayton, Sykes replied that their value was mainly negative. The
Arabs were ‘bad if they are against us, because they add to the enemy’s forces,
and if they are on our side there is so much less for the enemy and a little
more for us, but I do not like to count upon them as a positive force to us’.
To Balfour the situation was clear: ‘If we decide to do nothing, first of all
we shall lose the Sherif, and after him we shall lose the Arabs, and lose
them forever’. However, Lloyd George and Crewe – again deputising for
Grey – first wanted to know whether or not a landing at Alexandretta was
feasible, because, as Crewe argued: ‘it is no good starting on any proposals
with France until we have made up our mind that a big military effort is
possible’.
Before Sykes
withdrew, he was given the opportunity to emphasize once again that:
The question is very
urgent: it is important that a decision should be given quickly. Every day that
we delay we lose more and more Arabs from our side, and every day that we put
off brings us nearer to the day when there will be many Turks in Syria.
After he had left,
the members of the War Committee further discussed the Arab question. Kitchener
once again repeated that ‘the offensive-defensive’ – as Balfour put it – was
indeed the best way to defend Egypt. Balfour proposed that the French send troops,
although not to Alexandretta, but to Ayas Bay. Kitchener concurred,
as ‘the Turks expect us at Alexandretta, which has been entrenched, but there
are no entrenchments at Ayas’. Asquith believed that this was ‘an
attractive programme’. At the suggestion of Crewe, it was decided first to
consult Bertie before approaching the French government.²¹
Crewe dispatched a
letter to Bertie the next day. He acquainted the ambassador with the views,
Sykes had expressed before the War Committee. As far as the ‘offensive
defensive’ was concerned, Crewe fully realized that ‘then we come up against
French susceptibilities and claims, and any discussion becomes exceedingly
delicate, because the French always seem to talk as though Syria and even
Palestine were as completely theirs as Normandy’. The War Committee therefore,
believed that ‘it might be advisable for Mark Sykes to go over to Paris,
accompanied, perhaps, by someone like Fitzmaurice, in order to talk to some of
the French Ministers. He could press his own views upon them without committing
us to any particular movement.’ Bertie, however, opposed ‘the Sykes expedition
to Paris’. He argued that:
However intelligent
Sir Mark Sykes may be, and however good his arguments, I do not think that his
coming to Paris to talk to some of the French Ministers would be in the least
useful. However much he might press his views as his own views, they would be
regarded as the views of the British government; for otherwise, why should he
come?
The ambassador was
prepared to sound Briand personally on ‘a possible joint expedition somewhat
north of Egypt’, but warned that ‘contrary to Kitchener’s persistent
contention, they hold that Salonica and the possibility of an expedition
somewhere not defined will prevent the Germans starting any considerable
Turkish or Turco– German force for a march to Egypt’.²²
On 28 December 1915,
the War Committee indirectly decided to shelve the whole project. It accepted
the recommendations on military policy for 1916 made by Lieut.-General Sir
William Robertson, Murray’s successor as CIGS.²³ These were based on the
decision taken at an inter-allied conference at Chantilly on 8 December that
the war could only be won on the Russian, French and Italian fronts, and that
the number of troops on the other fronts should be reduced to the barest
minimum. Consequently, an ‘offensive-defensive’ policy for
the defence of Egypt was out of the question, at least for 1916. In
his memoirs, Hankey observed that:
Robertson must have
come away from the meeting of December 28th well satisfied. He had obtained the
adoption of his main principle that the western front was the main theatre of
war and he had been authorized to prepare for a great offensive there.
He had also secured the application of the principle of a defensive role to the
Egyptian and Mesopotamian campaigns.²⁴
Sir Mark Sykes And François Georges- Picot Come To An
Agreement
The second meeting
between the Nicolson Committee and Georges-Picot took place on 21 December
1915. Picot informed the British delegation that ‘after great difficulties, he
had obtained permission from his government to agree to the towns of Aleppo,
Hama, Homs, and Damascus being included in the Arab dominions to be
administered by the Arabs’.²⁵ The discussion then turned to the boundaries of
the area that should come under direct French administration, as well as the
question of which part of the future Arab state would fall within the French
sphere of influence. With respect to the latter, it was agreed that the Arab
state should be ‘divided between England and France into spheres of commercial
and administrative interest, the actual line of demarcation to be reserved, but
[…] that it should pivot on Deir el Zor eastward and westward’.
It was also decided that the Lebanon, which ‘should comprise Beirut and the
anti- Lebanon’, and an enclave around Jerusalem should be excluded from the Arab
territories.
Two points were
reserved for further discussion: ‘the allocation of the Mosul Vilayet [and] the
position of Haifa and Acre as an outlet for Great Britain on Mediterranean for
Mesopotamia’.²⁶ This fresh delay made Nicolson complain to Hardinge that
‘our discussions with the French in regard to the Arab negotiations are
proceeding exceedingly slowly, and I cannot say that I see much prospect of our
coming to an agreement’.²⁷ Sykes on the other hand, was rather sanguine. On 28
December, he in- formed Clayton that he had ‘been given the Picot negotiations.
I have prepared to concede Mosul and the land north of the lesser Zab if Haifa
and Acre are conceded to us.’ Sykes expected that it would not take him ‘above
3 weeks’ to solve the last problems with the ‘Picot negotiations’.²⁸
Sykes’s optimism
turned out to be justified. Within a week he came to an understanding with
Georges-Picot. The terms of the proposed agreement were laid down in a
memorandum that reached the Foreign Office on 5 January. Sykes and Picot
claimed that three parties were involved in a settlement of the Arab question –
France, the Arabs and Great Britain – and that each cherished territorial,
economic and political ambitions that could not be satisfied without coming
into conflict with those of the difficulties, he had obtained permission from
his government to agree to the towns of Aleppo, Hama, Homs, and Damascus being
included in the Arab dominions to be administered by the Arabs’.²⁵ The
discussion then turned to the boundaries of the area that should come under
direct French administration, as well as the question of which part of the
future Arab state would fall within the French sphere of influence. With
respect to the latter, it was agreed that the Arab state should be ‘divided be-
tween England and France into spheres of commercial and administrative
interest, the actual line of demarcation to be reserved, but […] that it should
pivot on Deir el Zor eastward and westward’. It was also decided
that the Lebanon, which ‘should comprise Beirut and the anti- Lebanon’, and an
enclave around Jerusalem should be excluded from the Arab territories.
Two points were
reserved for further discussion: ‘the allocation of the Mosul Vilayet [and] the
position of Haifa and Acre as an outlet for Great Britain on Mediterranean for
Mesopotamia’.²⁶ This fresh delay made Nicolson complain to Hardinge that
‘our discussions with the French in regard to the Arab negotiations are
proceeding exceedingly slowly, and I cannot say that I see much prospect of our
coming to an agreement’.²⁷ Sykes, on the other hand, was rather sanguine. On 28
December, he in- formed Clayton that he had ‘been given the Picot negotiations.
I have prepared to concede Mosul and the land north of the lesser Zab if Haifa
and Acre are conceded to us.’ Sykes expected that it would not take him ‘above
3 weeks’ to solve the last problems with the ‘Picot negotiations’.²⁸
Sykes’s optimism
turned out to be justified. Within a week he came to an understanding with
Georges-Picot. The terms of the proposed agreement were laid down in a
memorandum that reached the Foreign Office on 5 January. Sykes and Picot
claimed that three parties were involved in a settlement of the Arab question –
France, the Arabs and Great Britain – and that each cherished territorial,
economic and political ambitions that could not be satisfied without coming
into conflict with those of the other two. From this it followed that ‘to
arrive at a satisfactory settlement, the three principal parties must ob-
serve a spirit of compromise’. This settlement would, moreover, have ‘to be
worked in with an arrangement satisfactory to the conscientious desires of
Christianity, Judaism, and Mahommedanism regarding the status of
Jerusalem and the neighboring shrines’. In the light of these
considerations, they had arrived at the following proposal (see Sykes-Picot map
of 1916):
1. Arabs. – That
France and Great Britain should be prepared to recognize and protect
a confederation of Arab States in the areas (a) and (b) under the suzerainty of
an Arabian chief. That in area (a) France, and in area (b) Great Britain,
should have priority of right of enterprise and local loans. That in area (a)
France, and in area (b) Great Britain, should alone supply advisers or foreign
functionaries at the request of the Arab confederation.
2. Great Britain,
should be allowed to establish such direct or indirect administration or
control as they desire.
3. That in the brown
area [which covered the greater part of Palestine; R.H.L] there should be
established an international administration, the form of which is to be decided
upon after consultation with Russia, and subsequently in consultation with Russia,
Italy, and the representatives of Islam.
4. That Great Britain
be accorded (1) the ports of Haifa and Acre, (2) guarantee of a given supply of
water from area (a) for irrigation in area (b). (3) That an agreement be made
between France and Great Britain regarding the commercial status of Alexandretta,
and the construction of a railway connecting Bagdad with Alexandretta.
5. That Great Britain
have the right to build, administer, and be sole owner of a railway connecting
Haifa or Acre with area (b), and that Great Britain should have a perpetual
right to transport troops 2. That in the blue area France, and in the
red area along such a line at all times.
On the same day, Nicolson
circulated copies of the memorandum to Holderness, Brigadier-General
George Macdonogh, director of military intelligence (DMI), and Captain
Hall, director of the intelligence division (DID) at the Admiralty. In his
covering letter he stated that, although ‘of course the agreement merely
represents the personal views of Sir Mark Sykes and M. Picot’, he believed that
it presented ‘a fair solution of the problem’.²⁹ Only the India Office agreed
with Nicolson’s conclusion. The loss of Mosul would clearly be ‘a serious
sacrifice for us’, but, on the other hand, it would force the French ‘to be
very accommodating elsewhere, e.g. Haifa’. The India Office should like to see
some modifications in the proposed terms, but on the whole the memorandum, as Hirtzel noted,
‘represents a considerable abatement on M. Picot’s original claim, and we are
under a great obligation to Sir Mark Sykes’.³⁰ Macdonogh and Hall
were considerably more critical. They accepted that an early settlement of the
Arab question was important to prevent a jihad, but questioned the assumption
that an agreement with France had to be reached first, before the Arabs could
be dealt with. Macdonogh argued that:
To me it appears that
the one point of importance is to get the Arabs in on our side as early as
possible. I would therefore, suggest that all that is necessary at the moment
is that we should be in a position to inform the Sheikh what are the approximate
limits of the country which we and the French propose to let him rule over.
This may involve an agreement as to the respective British and French spheres
of influence in that district, but I hope that its discussion will not be
allowed to delay the settlement of the main question.³¹
Hall, for his part,
doubted whether it was ‘necessary to have some agreement with the French about
Syria and Mesopotamia, in order that such action may be taken as may avert a
combination between the Turco–German forces and the Arabs, the result of which
would produce something like a serious general Moslem jehad against us’.
According to the DID, ‘action, which will convince the Arabs of our effective
power, is very necessary’. In- deed, ‘force is the best Arab propaganda’, and
it was therefore very desirable that a concerted naval or military action be
undertaken that would ‘result in cutting off the Arabs from the Turks by an
occupying force and so screening the former’, but precisely ‘no such action on
the part of the French, or on our part with their good-will and furtherance is
a term of the agreement’. The proposed agreement was moreover unsatisfactory
considering the assurances Hussein had asked for:
(a) That
the Arabs shall not be deserted by the Allies in any peace which may be made;
and
(b) That
all territories properly considered as inhabited by Arabs shall (with certain
exceptions) be part of an independent Arab State, guaranteed by the Allies. He
does not appear ever to have been willing to exclude Syria, and more especially
the Arab center of Beirut, from the Arab State.
Further, he and other
Arab leaders in touch with the British have, on several occasions expressed
themselves very emphatically against their being placed under any obligation to
accept French advisers locally, whereas they stated that they were prepared to
welcome British.
These considerations
led Hall to the conclusion that ‘the only advantage’ of the proposed agreement
that ‘would at present be gained seems to me the possibility of giving
definiteness to the assurances which would in them- selves be
unsatisfactory’.³² Finally, both Macdonogh and Hall could not help
thinking that, as the former put it, ‘we are rather in the position of the
hunters who divided up the skin of the bear before they had killed it’.³³
Pending Picot’s return from Paris, the observations by Hirtzel, Macdonogh and
Hall drew no comments from Grey or his officials.
On 16 January 1916,
Sykes informed the Foreign Office that he had spoken to Picot, and that the
latter had informed him that ‘at Paris he had much difficulty, but that he
believed that it would be possible to come to an agreement on the lines of the
memorandum’.³⁴ Nicolson convened a further meeting of the interdepartmental
committee on 21 January. During that meeting, ‘the criticisms of the various
Departments on the Sykes-Picot Memorandum were considered and no insurmountable
difficulty to the scheme was put forward in any of them’. Nicolson impressed
upon the other delegates that it was ‘essential to take France in our
confidence before we embarked on final negotiations with the Arabs’, and it was
again laid down that ‘if the Arab scheme fails the whole scheme will also fail
and the French and British governments would then be free to make any new
claims’. Sykes was authorized to inform Picot of the results of the meeting, as
well as that ‘H.M.G. would feel compelled to consult the Russian government after
agreement with the French’ on the northern frontier of the blue area.
As a result of this
meeting, the Foreign Office drew up a draft agreement. Its conditional
character was emphasized by adding a preamble stating that ‘should the
negotiations with the Grand Shereef of Mecca fail to secure the active
cooperation of the Arabs on the side of the Allies the whole proposals in
regard to all spheres whether of administration or of influence will lapse
automatically’. The India Office and the DMI concurred with the draft
agreement. Holderness commented that it was ‘in accordance with the conclusions
reached by the Committee on Friday’, while Macdonogh ‘quite agree[d]
with its contents’. Hall, however, protested anew against the absence of a
‘stipulation for French cooperation in, or consent to, any concerted plan of
action against the Germans and Turks as a condition of the agreement’.³⁵
The Foreign Office
completed the final draft on 2 February. It was circulated to the Cabinet that
same evening, with a covering letter in which Nicolson explained the reasons
for negotiating this agreement with France. It had been ‘found at the outset impossible
to discuss the northern limits of the future Arab State or Arab Confederation,
unless the French desiderata in Syria were also examined, as M. Picot was
unable to separate the two questions’. Eventually, it had been agreed that ‘the
four towns of Homs, Hama, Aleppo and Damascus will be included in the Arab
State or Confederation, though in the area where the French will have priority
of enterprise, etc’. Nicolson did not fail to point out that the preamble
was intended to lay down ‘with sufficient precision’ that ‘the proposals in
regard to the Blue area, as well as the Red area are contingent on the
fulfilment of certain essential conditions’, and that Russia should be given
full opportunity to have a say in the final settlement of the question.³⁶
The War Committee
considered the matter the following day. It was decided on the suggestion of
Sir Ed- ward that ‘the whole Arab Question should be discussed at a meeting
between Mr Bonar Law [the secretary of state for the
colonies], Mr Chamberlain, and Lord Kitchener, and that the French
should be informed if we agreed to their proposals’.³⁷ This meeting took place
the next day. Crewe and Nicolson were present, as well as Holderness
and Hirtzel; ‘a representative of the Admiralty was also present, but was
not in a position to give an opinion on the merits of the scheme’. Those who
were decided that:
M. Picot may inform
his government that the acceptance of the whole project would entail the
abdication of considerable British interests, but provided that the cooperation
of the Arabs is secured, and that the Arabs fulfil the conditions and obtain
the towns of Homs, Hama, Damascus and Aleppo, the British government would not
object to the arrangement. But, as the Blue Area extends so far eastwards, and
affects Russian interests, it would be absolutely essential that, before
anything was concluded, the consent of Russia was obtained.
On the evening of 4
February, Sir Arthur informed Georges-Picot of the British decision.
He minuted afterwards that he had laid ‘emphatic stress on the
absolute necessity of nothing whatever being considered settled until the
Russian consent had been obtained – and […] that we should say nothing to the
Arabs until that consent has been obtained’.³⁸ Five days later, Cambon told
Nicolson that ‘the French government are in accord with the proposals
concerning the Arab question’.³⁹ Sykes and Picot were entrusted with the task
to inform the Russian authorities of the contents of the agreement.
The Arab Question Becomes A Regular Quicksand
Foreign Office
officials had little time to savor the successful conclusion of the
negotiations with the French on the Arab question. On 5 February, Oliphant and
Nicolson occupied themselves with Hussein’s reply to McMahon’s letter of 14
December 1915 (see section ‘Georges-Picot’s Opening Bid and McMahon’s Third
Letter’, above). This letter, dated 1 January 1916, had been received by the
Foreign Office on 2 February. The high commissioner had declared in a telegram
of 26 January that the letter was ‘of friendly and satisfactory nature’,⁴⁰ but
after they had studied it both Oliphant and Nicolson disagreed. The
former minuted that he could not ‘regard
the Sharif’s letter as very satisfactory, though it is at least
outspoken and frank’, while the latter observed that he did not:
Consider this letter
at all satisfactory as regards the Sharif’s remarks respecting the
French and I wish in his telegram […] Sir H. McMahon had given us some
indication of this – He made no mention of the northern parts in his telegram –
and we have had to believe that the Shereef had not taken serious notice of
them while on the contrary he employs rather ominous language in regard to
them.
About the Emir’s
position on these ‘northern parts,' McMahon explained in his covering dispatch
that:
Satisfactory as it
may be to note his general acceptance for the time being of the proposed
relations of France with Arabia, his reference to the future of those relations
adumbrates a source of trouble which it will be wise not to ignore.
I have on more than
one occasion brought to the notice of His Majesty’s Government the deep
antipathy with which the Arabs regard the prospect of French Administration of
any portion of Arab territory. In this lies considerable danger to our future
relations with France, because difficult and even impossible though it may be
to convince France of her mistake, if we do not endeavour to do so by
warning her of the real state of Arab feeling, we may hereafter be accused of
instigating or encouraging the opposition to the French, which the Arabs now
threaten and will assuredly give.⁴¹
McMahon’s
observations reflected Clayton’s anxieties, which the latter had voiced in two
letters to Wingate. The Sudan agent considered ‘the Sharif’s answer
[…] on the whole satisfactory’, but taken together with the results of the
second meeting with Picot on 21 December, he feared that the British could not
go on ‘negotiating much longer, without laying ourselves open to a charge of
breach of faith, unless we honestly tell the Arabs that we have made Syria over
to the French’. A problem that was the more important since:
Some of our Syrian
friends seem to have an inkling that we have handed Syria over to the French
and I foresee some trouble. The time has nearly arrived when we shall have to
tell them so straight out and hand them over to the French to settle with –
other- wise we shall risk giving rise to the very friction with France that we
have sacrificed so much to avoid.⁴²
Wingate was more
optimistic. According to him the results of the meeting were: On the whole not
quite so unsatisfactory as I had expected, and I think I see in the general
trend of the discussion, the possibility of coming to an arrangement which may
satisfy all parties – indeed I do not see that even if French demands are
conceded in their entirety, that we can be accused of any serious breach of
faith – it is true the Arabs will not get all they wanted, but they will
achieve a great deal and in any circumstances, I should think that further
discussions will result in a certain modification of the French demand.⁴³
However, Clayton, in
a further letter, confessed that he did not ‘share these hopes’. He enclosed
copies of McMahon’s covering dispatch with Hussein’s fourth letter, and the
reply the latter intended to send ‘without waiting for formal approval’.
Clayton explained that it had not been an easy assignment ‘having only a couple
of hours to do it in, and [having] to steer clear of the various quick
sands and yet to say something which would satisfy the Sharif’.⁴⁴
Clayton could have spared himself the trouble as far as Grey was concerned. To
him, the Arab question already was ‘a regular quicksand’.⁴⁵
Cambon was rather
more sanguine. In view of McMahon’s suggestion to warn the French ‘of the real
state of Arab feeling’, the India Office had expressed the desire that the
Foreign Office should do so. Of course, it was ‘not unlikely that they will not
take the statement seriously. But Chamberlain apprehends that His
Majesty’s Government may hereafter be under some suspicion of bad faith if,
with the information before them, they allow the negotiations to proceed
without warning the other party.’ Grey had consequently instructed the
department to mention the matter to Cambon.⁴⁶ As the India Office had
predicted, the latter did not take the matter very seriously. He cheerfully
remarked to Nicolson that ‘the Shereef would not be an Arab if he did not say
something of that kind’.⁴⁷
Sykes, meanwhile,
acted as advisor to the British ambassador at Petrograd, Sir George Buchanan,
during the latter’s negotiations on the frontiers of the blue area with the
Russian minister for foreign affairs, Sazonov, and the French ambassador, Paléologue,
who was assisted by Georges-Picot. Grey had observed in his instructions that
Britain had ‘no desire whatever to urge the Russian government to make
concessions in the districts which are of direct interest to them if they have
any objections to doing so’.⁴⁸ Sazonov indeed objected. At the first
meeting of the three parties he showed ‘very plainly he did not like extension
of the blue area so far eastward’.⁴⁹ However, a compromise was reached within
two days, to the effect that the most eastern part of the blue area would
become part of the area under direct Russian administration, while France would
be compensated for the loss of this region ‘by enlarging her blue area to the
north of Marash’.⁵⁰
On 17 March, Buchanan
telegraphed that the Russian government had decided to accept the compromise.⁵¹
At a meeting of the War Committee six days later, it appeared that Balfour,
Kitchener, and Asquith objected to the proposed scheme, albeit on different
grounds. Each time, Grey tried to neutralize their objections
by emphasizing that ‘the whole arrangement was provisional on the
Arabs coming in. Unless they did, there would be no break up of Asia Minor,’
and that, accordingly, ‘he thought that nothing would come of all this, [and]
Asia Minor would never be divided’.⁵² Eventually, it was decided that ‘His Majesty’s
Government would raise no objections to the proposed arrangement between France
and Russia’.⁵³
Despite this
progress, negotiations again could not be brought to a conclusion. Fresh
problems arose with respect to ‘all concessions for railway construction and
other advantages such as religious missions granted to the French by the Turks
in any territory that Russia may acquire’.⁵⁴ The result was that, on 3 April
1916, 171 days after Maxwell had telegraphed that ‘time is of the greatest
importance, and that unless we make definite and agreeable proposal to the
Shereef at once, we may have a united Islam against us’, and 110 days after
Sykes had testified before the War Committee that ‘the question is very urgent:
it is important that a decision should be given quickly. Every day that we
delay we lose more and more Arabs from our side,’ Buchanan still had to impress
on Sazonov the importance of a speedy conclusion of the negotiations
in order that Britain would be ‘able to clinch matters with Arabs at once’.⁵⁵
Sir Mark had
indicated some weeks before that two potential dangers threatened the Arab
revolt:
‘1. Peninsula nomads
moving before intellectual Syrians are prepared and scheme failing through want
of organisation.
2. Of
intellectual Syrians failing to combine with intellectual Mosul
and Irak Arabs to join in movement owing to doubt as to our designs
on Irak’. Concerning the latter, Sykes had suggested sending ‘Arab and
Kurd officers now Turkish prisoners of war in India to Egypt and letting
Colonel Clayton sound those committed to the Arab cause and select best to work
with Masri and Faruki’. Although Oliphant
had minuted that he could not ‘conceal my skepticism as to
the success of the scheme’, Sykes’s telegram had been repeated to Cairo. The
next day, yet an- other telegram had been sent to McMahon, in which he had been
informed that ‘no action whatever should be taken on it’ (i.e. Sykes’s
telegram), but that the Foreign Office would ‘be glad of your observations on
it’.⁵⁶ McMahon considered it wiser to send Aziz Ali and Faruqi to
Mesopotamia and there to get in touch with ‘the Arab element in the Turkish
Army’. There was, however, the problem that they:
Demand for themselves
and Arab military element whom they would have to approach some definite
assurance of British policy towards Arabia. They consider this essential to the
success of any effort to win over Arab element in the Army.
They would be
tolerably content with the assurances already given to the Shereef. Their
tendency at present is to demand less from us with regard to Mesopotamia than
would have been acceptable before.
Oliphant supported
McMahon’s ‘suggestion that these two men should go’. Grey and Kitchener, too,
were in favor of the proposal. Together they drafted – ‘at the
Cabinet this morning’ – a telegram in which the high commissioner
was authorized, provided Clayton did not object, to
send Faruqi and Aziz Ali to Mesopotamia. They also permitted Sir
Henry ‘to give assurances, if necessary, but you should be very careful not to
exceed in any way the limits of the assurances already given to the Shereef’.⁵⁷
Copies of both
telegrams were forwarded to the India Office. Chamberlain was not amused. The
India Office drew the Foreign Office’s attention to Husayn’s letter
of July 1915, in which the Emir:
Purported to speak
for the ‘Arab Kingdom of the Shereef’, while in that of 1st January he
expressly stated that his procedure was not personal, but the result of the
decisions and desires of his peoples of which he was only the transmitter and
executant. There is no clear evidence as to how far this claim accords with
facts, but it has not, so far as Chamberlain is aware, been
questioned by His Majesty’s Government. If the claim is well founded, it is a
point for consideration whether independent assurances should be given to
others, and ex hypothesize less responsible Arabs.
Oliphant minuted that
the telegram to Sir Henry ‘was not a departmental draft,' and proceeded to
draft a telegram in the sense of the India Office letter. Nicolson was
embarrassed by the letter, although he ‘understood that Sir E. Grey and Lord
Kitchener consulted M. Chamberlain before the telegram was dispatched’. He
admitted that there was a good deal of force in the concluding remarks of the
I.O. letter’, but the text of Oliphant’s draft ‘rather clashes with the
telegram sent in 54229 [the one drawn up by Grey and Kitchener] – and would
possibly confuse Sir H. McMahon’.⁵⁸
Two days later, the
Foreign Office received another letter from the India Office, enclosing a
telegram from General Sir Percy Lake, GOC-in-C, Force ‘D’. The latter was
opposed to McMahon’s suggestion. It was:
Not considered
possible that either of the above individuals could themselves pass over from
occupied territory to the sphere of the Turkish troops op- posed to us on the
Tigris or Euphrates, or could be of any practical use to us if they did. From
the political standpoint it appears to us that their political views and
schemes are much too advanced to be safe pabula for the communities
of occupied territories and their presence in any of the towns
of Iraq would be in our opinion undesirable and inconvenient.⁵⁹
Lake’s telegram had
been repeated to Cairo and McMahon promptly reacted. He explained that ‘it was
not intended that Al Masri and others should pass over to Turkish
lines’. All that had been envisaged was that ‘presence of one or two prominent and
carefully selected members of the Arab party in our ranks would afford Arab
elements in Turkish army much required guarantee of our unity of interest and
good faith’. He moreover warned that the decision not to send Aziz Ali
and Faruqi would:
Produce
disappointment and rumors of danger being ascribed either to our
mistrust in their loyalty, or to our unwillingness, if not inability, to carry
out our assurances, and this may not be without effect on Shereef. An
impression is gained that there is a visible limit to the patience of those in
whom we have raised feelings of expectation nor is it possible to guarantee
that the present favorable attitude of certain individuals can be
counted on later.
McMahon, therefore,
trusted that he might ‘continue to give all guarantees short of definite action
and within the limits approved by you to those who have now committed their
destinies to us.'
This telegram induced
Chamberlain to compose a very biting memorandum:
I do not find this
telegram very easy to understand.
The decision to which
it refers is that El Faruki and El-Masri should proceed to
Mesopotamia. As it now appears that Sir H. McMahon never contemplated that they
should pass over to the Turkish lines (as was supposed here), it is not clear of
what use he thought they could be. It is not believed that either of them have
any influence in Irak. How is ‘practical use’ to be made of them?
‘An impression is
gained’, Sir Henry telegraphs, ‘that there is visible limit to the patience of
those in whom we have raised feelings of expectation’. This is the severest
criticism I have seen of Sir H. McMahon’s policy. He raised the expectations.
We have given assurances by his mouth much wider than we at home intended: We
have given money and arms and promised more. The Sharif has done
nothing, and we are now to be told by Sir H. McMahon that it is we who fail to
fulfil the expectations we have raised! Will Sir Henry
ever realize that there are two sides to a bargain that the Shereef
has his part to play and that it is now ‘up to’ him the Shereef to make the
next move?
What does he mean by
‘continuing to give all guarantees short of definitive action?’ He has given
guarantees as already stated in excess of our intentions. He safeguarded French
freedom of action in Syria but not ours in Mesopotamia. But by his declarations
we hold ourselves bound and there has been no suggestion that we should recede
from them. If he only desires to repeat himself, he has authority to do so, but
does he mean that he is to give further assurances, and if so what? I am very
uneasy about the whole handling of the question by Egypt.
Grey’s reaction to
Chamberlain’s complaint was very characteristic. He did not enter into a
discussion on the merits of the latter’s arguments. He confined himself to a
brief note to Nicolson:
You will see
what Mr Chamberlain says. I am disposed simply to telegraph to Sir H.
McMahon that I do not understand his difficulty about assurances that he can
repeat assurances already given but must not go beyond them, that we are I
believe giving arms and money and the sole question is whether and when the
Arabs will do their part.
A telegram in this
sense was dispatched to Cairo on 5 April 1916.⁶⁰
Sykes, the man whose
suggestion had started this controversy, had in the meantime returned to
London.
There he set himself
to solving the problem that according to McMahon constituted the biggest threat
to a satisfactory solution to the Arab question, ‘the deep antipathy with which
the Arabs regard the prospect of French administration of any portion of Arab
territory’. In a telegram sent from Petrograd on 16 March, Sykes had already
declared that ‘with regard to Arabs our greatest danger lies in their falling
out with the French’, but that ‘if I can get Picot and Faroki or Aziz
Ali into a room together, I believe I can manage to patch up a bargain between
them’. He had therefore advised:
Get
El Masri or Faruki or both to London where I could enter
into formal discussion with them and when ground was prepared bring them into
contact with Picot. I suggest this as I fear French and Arab discussions in
Cairo leading to intrigues and quarrels and Picot would like this arrangement.
If Arabs reach London April 7th, I believe by May 8th ground would be clear of
Arab French question.⁶¹
In spite of this
hopeful prospect, the Foreign Office had not acted on Sykes’s suggestion. This
did not prevent Sir Mark from submitting to the Foreign Office a telegram to
Clayton in which he asked the latter’s opinion on this scheme. After
consultations with Macdonogh, the telegram was finally sent off on 14
April. Sykes first informed Clayton on the situation with respect to the Anglo–
French–Russian negotiations in Petrograd, as well as the compromise that had
been reached as to the limits of the blue area, and continued:
The crux of the difficulty
is that at present French theoretically concede no outlet to Arab State on
Syrian littoral. They intend to negotiate this point with Arabs themselves.
Negotiations on this point through any medium in Cairo will precipitate the
Maronite versus Anglophil controversy. In my mind it is essential
that French should have become practical before Picot goes to Egypt. I advise
therefore following procedure which I have got Picots approval of, i.e. that
you send here to London 2 Arab officers representative of intellectual Syrian
Moslem Arab mind, that when I have got their point of view that I compare it
with Picots, that when Picot has been got into right frame of mind I bring them
together and they have informal talk, Picot then gets Paris to make concession
of principle of Arab State outlet on Syrian littoral in the form of
Aide-Memoire to H.M.G. […] Objection was taken to this procedure in London on
ground that Arab officers would not be representative and that negotiations
would be being conducted in two places at once. I wish to make it clear that
suggestion is not to negotiate but to examine, and that official status of Arab
officers is not important as long as they are mentally representative.⁶²
Unfortunately for
Sykes, Clayton was opposed to the scheme. However much he agreed with Sir Mark
that Pi- cot’s presence in Egypt would be undesirable at the present juncture,
he did not see how ‘presence in London of any Arab officers with whom we are at
present in touch would in any way assist you.' According to Clayton, it would
moreover ‘be most impolitic to raise now with Arabs Syrian question which is
quiescent for the moment. To do so would, I am convinced, be contrary to
interests both of ourselves and French, who have everything to gain by delay.’
There was the added problem that at the moment there
seemed to take place a ‘certain rapprochement’ between the Arabs and the
Turkish opposition (the ‘Turkish decentralization parties’, as
Clayton called them) – a development that McMahon would explain in a further
telegram – and this meant that ‘an attractive offer by such Turkish parties
would be in serious competition to any proposals Allies can put forward’.
Against this back- ground, Clayton also did not think it very advisable to
disclose the results of the Anglo–French–Russian negotiations on the future of
the Ottoman Empire. Of course:
Any agreement on main
principles between Allies is all to the good, but to divulge it at present and
to insist on any particular program Anglo-French would I am
convinced, raise considerable feeling, to strengthen Arab-Turkish
rapprochement, and possibly to affect the injuriously political and military
situation of Allies in Turkey at a moment when true attitude of Arabs is not
quite clear.⁶³
As Clayton’s telegram
was addressed to Sykes, Foreign Office officials did not comment on it, but
they did, and in very strong terms, comment on McMahon’s telegram when it
arrived two days later. The high commissioner repeated Clayton’s objections to
raising the Syrian question at the present moment. He also confessed that he
was ‘unaware whether proposals outlined by Sir M. Sykes have received the
approval of three governments concerned or whether they are merely suggestions
as a result of his and Picot’s conversations in Petrograd’. McMahon hoped that
the latter was the case, as there were indications that ‘Turkish parties in
opposition to Committee of Union and Progress are already considering peace
terms which, in certain circumstances, it might suit the Allies to consider’.
The Petrograd compromise moreover appeared ‘to ignore existence of Turkey and
necessity, under any circumstances of providing an adequate home for remnants
of that nation if defeated’.
In their minutes,
Oliphant and Nicolson gave vent to their feelings of frustration with the
manner in which the Arab question developed. Oliphant’s minute could have been
written by Hirtzel or Chamberlain: I venture to think that
this telegram is by no means satisfactory. It shows that there is considerable
confusion in Sir H. McMahon’s mind and that matters are merely drifting […]
1.To state that the
future of the Turks is ignored is erroneous […] The Vilayets of Brusa,
Smyrna, Angora, Konia, Kustammi and Eskisher – an area as
large as France and the only districts inhabited by an Ottoman Turk majority
are not touched by the agreement in question, which were from the outset
drafted on the basis of ethnographic interests.
2. The arrangements were devised to fall in
with Turkish liberal views if the Turkish liberals were strong enough to oust
the C.U.P. The Turkish liberals if in power, would have to approach Russia in
the final instance. As regards Turkish parties we know Sharif Pasha
and Saba ed Din are in Paris and some may be in Egypt [Sharif Pasha and
Prince Sabah-al-Din acted as spokesmen for the Turkish opposition; ].
But Enver is in Constantinople and the Sheed-Dirif in
Mecca.
It seemed to Nicolson
that there was no end to fresh complications. He minuted that he was
‘afraid that the whole subject is becoming entangled’, and that at the very
moment that ‘we, Russia and France are now quite clear and in accord as to our
interests and aspirations in the Ottoman Empire’. As far as the proposals of
the Turkish opposition were concerned, these were ‘merely empty talk’. The
permanent under-secretary concluded that the Foreign Office should ‘let Sir H.
McMahon fully know the present position and our arrangements with Russia and
France as regards Asiatic Turkey and also inform him of our attitude towards
the irresponsible and unofficial overtures made to us by
Prince Sabadeddin and others’. Crewe agreed, and a telegram in the sense
of Oliphant’s and Nicolson’s minutes was sent to Cairo on 27 April.⁶⁴
In his reply of 4
May, McMahon voiced his disappointment with the Foreign Office attitude towards
the Turkish ‘liberal and anti-committee parties’, especially as the ‘situation
as far as we can gauge it here, does not yet appear one in which we can afford
to disregard potential value of this disintegrating factor in Turkey’. He also
observed that:
Although there is
nothing in arrangement agreed on between France and Russia and ourselves as
defined in your telegram that conflicts with any agreements made with or
assurances given to Shereef and other Arab parties, I am of opinion it would be
better if possible not to divulge details of that arrangement to Arab parties
at present.
Moment has not yet
arrived when we can safely do so without some risk of possible
misinterpretation by Arabs.
Grey personally
drafted a telegram to Cairo in which he agreed that ‘details of arrangement
should not be divulged’.⁶⁵
In his minute on Sir
Henry’s telegram of 22 April, Oliphant had also submitted that in order:
To avoid further
confusion […] a meeting between Sir M. Sykes, M. Picot, and Col. Clayton would
be helpful not in Egypt as M. Picot’s presence there is obviously inopportune.
But perhaps at Rome or even Paris: at any rate somewhere within reach of London.
So that eventually
decision can be derived at here, and the details worked out at Cairo.⁶⁶
On 3 May, Clerk reported that Sykes was
in favor of a meeting between him, Picot and Clayton, perhaps at Paris.
Nicolson saw no objection, but Kitchener doubted whether Clayton could be
spared from Egypt.⁶⁷ This seemed to be the end of Oliphant’s suggestion, but in
a memorandum on a conversation with Georges-Picot, who had returned from
Russia, Sykes again raised the subject. He reported that the French and Russian
governments had settled their part of the arrangement on the Middle East by
means of a mutual exchange of letters be-
tween Paléologue and Sazonov, and that the French government
wished that Grey and Cambon should follow the same procedure. Sykes urged the
importance of an early exchange of these letters, also because ‘exchange of
notes is an essential prelude to a conference between M. Georges-Picot and
Colonel Clayton on Franco–British Arab policy. Such a conference M.
Georges-Picot earnestly desires.’ ⁶⁸
Despite Sykes’s plea for a speedy settlement, an
exchange of letters between Grey and Cambon did not take place immediately.
Grey first wanted to make sure whether the French were fully aware of ‘the
point of its being conditional upon action taken by the Arabs’. During an
interview with Grey, Cambon assured him that ‘it was well understood that it
was dependent upon an agreement with the Sharif of
Mecca and that this provisional character was already in writing’.⁶⁹ There
was also the point whether or not the French government, in those areas that
would ‘become entirely French, or in which French interests
are recognized as predominant’, would respect ‘any existing British
concessions, rights of navigation or development, and the rights and privileges
of any British religious, scholastic or medical institutions’. A letter to this
effect was sent to the French ambassador on 15 May. In his reply, which reached
the Foreign Office the following day, Cambon confirmed that France would
maintain existing British rights, privileges, and concessions, whereupon Grey
dispatched a letter to the French ambassador that same day that contained the
terms of what would go down in history as the Sykes– Picot agreement.⁷⁰
Now that this
obstacle was out of the way, Clerk once again pressed for a meeting between
Picot, Sykes and Clayton ‘as soon as possible’.⁷¹ McMahon was not averse to the
proposed conference, but found it ‘extremely inconvenient to General Clayton
until the return of members of Arab Bureau and Storrs which should be in
ten days time.' It was also ‘desirable that Clayton should be able to take
home “first hand” information with regard to result of their mission’.⁷²
Ronald Storrs,
accompanied by Captain Kinahan Cornwallis and Commander David
Hogarth, had left Cairo for the Hijaz. The object of their mission was to meet
Abdullah, at the latter’s request. However, Abdullah was unable to make it to
the rendezvous. On 6 June 1916, the British delegation instead spoke
with Zeid, the youngest of Husayn’s sons. In his report, Hogarth
observed that the British delegates had feared that ‘the substitution
of Zeid for Abdullah had rendered it unlikely that […] we should be
in a position to appreciate the actual situation and future policy of the
Sherif’,⁷³ but they were in for a surprise. Zeid informed the British
delegation that the day before, Ali and Feysal, Husayn’s two
other sons, had started hostilities against the Turkish garrison at Medina. The
revolt of Sharif Husayn, Emir of Mecca, against his Turkish masters had
begun.
Preliminaries To Sharif Hussein’s Revolt
It was not so much
the fact that Hussein revolted that came as a surprise to the British
authorities at Cairo and Khartoum, as the moment he chose to do so. On 16
February 1916, Hussein had sent a letter to McMahon in which he not only
unfolded his plan of action but also asked for arms and money. Oliphant and
Nicolson hesitated to grant these demands, because, as Oliphant argued, to do
so might lead the Emir to think that Great Britain considered herself to be
definitely committed to him. Sir Arthur agreed. He was:
Anxious lest the
Shereef should consider by our sending him the additional £20,000 he may regard
his agreement with us as definitely concluded, and that we are bound to meet
all his desiderata. I consider that Sir H. McMahon should make it quite clear
to the Shereef that we are providing him with the sums for which he has asked
as an evidence of our friendly feelings towards him and that we let him know
later when we consider that the opportune moment has arrived for his taking
[undecipherable] action as will lead to the revolt of the Arabs against Turkish
rule – and which will result in the discomfiture of the Ottoman Empire in the
regions where Arab interests are predominant and where it is desired to
establish Arab independence – we should telegraph Sir H. McMahon in above
sense.
Lord Kitchener might
be shown this before it is dispatched – and I.O. concurrence obtained.
Neither Chamberlain
nor Kitchener had a good word to say for the proposed telegram. The former
scathingly minuted:
I have not thought it
very probable that the Grand Sheriff would take any definite action, Sir H.
McMahon seemed to me to have succeeded in giving him the impression that we
were in much more need of him than he of us.
However, now the
unexpected happens. The Sheriff declares that the time for action has come –
and we propose to wet blanket his enthusiasm! Subject to military opinion […] I
should pay the money and encourage him to go ahead.
We have warned France
that she must obtain the acquiescence of Russia. We have warned the Sheriff
that we cannot speak for our allies. What more has either of them a right to
expect from us and why should we discourage a potential ally?
I am not sure of the
Sheriff’s good faith, but at least we can now test it by taking him at his
word.
Kitchener concurred.
He had ‘no idea there would be any hesitation in paying the money and accepting
the Sharif’s proposed help. I hope the proposed telegram may be
modified and sent without delay.’ Nicolson rather timidly minuted that
‘given their opinion, I am afraid that we can only telegraph to Sir H. McMahon
to send the money and accept the Sharif’s proposals’. A telegram in
this sense was dispatched the same day.⁷⁴
Wingate regarded this
telegram as a signal success, witness his letter to McMahon of 17 March in
which he confirmed the receipt of the latter’s telegram:
Giving me the welcome
news that the British government has finally approved of
the Sharif’s demands and that everything is ‘en train’ […] I am
full of hope that the Arab Policy that we initiated so long ago is really going
to materialize. Of course, there is naturally some risk that things may
not go quite as we hope and expect they will, but in this, as in most
operations of the sort, it is a case of ‘Nothing venture nothing have’ and in
this matter especially I think the game is well worth the candle.⁷⁵
McMahon reported in a
telegram of 21 March that a messenger from Hussein had arrived at Port Sudan,
carrying yet another letter. Cyril Wilson had telegraphed a summary to Cairo.
On the basis of this summary, Sir Henry drew the conclusion that the ‘Shereef
appears to have made up his mind definitively to side actively with us and his
last two letters make no further reference to political matters so that he
would seem to be satisfied with assurances we have given him and to require
nothing further in this respect’, and that Hussein had decided that ‘rising
will begin at end of coming Arabic month or at beginning of month after’. At
the Foreign Office, this news impressed no one. Oliphant minuted that
McMahon’s telegram struck him as ‘very optimistic’, while other officials, as
well as Grey, merely initialed it.⁷⁶
On 24 May, Sir Henry
sent a telegram to the Foreign Office in which he reported that ‘Storrs is
urgently required by Abdullah Shereef’s son to meet him on Arabian coast.
Shereef is asking for £50,000 and Abdullah for £10,000.’ He strongly
recommended complying with these requests without delay ‘as (?matter) appears
to have reached point at which we must not fail to give it every
encouragement’. Clerk’s reaction showed that the various quick
sands and entanglements that seemed to inhere in the Arab question had
also not failed to make an impression on him. He minuted that ‘this
may be encouraging, but there is another interpretation, which is that fair
words cost nothing and are well worth £60,000’. He suggested:
To
send Mr Storrs across at once to hear what Abdulla has to say;
meanwhile collect the money at Port Sudan, and if Mr Storrs’ report
is favourable, hand it over. Mr Storrs can always explain that
it takes a little time to get £60,000 together, and that he has come across in
advance, so as to lose no time.
It took five days
before a telegram in this sense was sent to Cairo.⁷⁷ On the evening of 28 May,
without waiting any longer for Foreign Office authorization, Storrs, and
his companions left for the Hijaz. McMahon telegraphed that day that ‘they will
proceed to Port Soudan, pick up Sharif’s messenger and go to Hedjaz
coast to meet Abdullah. They take £10,000 for latter but sending of £50,000
for Sharif must await your sanction’. The Foreign Office acquiesced
to the high commissioner’s decision to send the £10,000, while the ‘payment of
£50,000 to Sharif is sanctioned if there is a real rising’.⁷⁸
A telegram from
McMahon transmitting Storrs’s report of his meeting with Zeid arrived
in London on 9 June. It stated that ‘rising began yesterday [5 June 1916] at
Medina but all communications in Hedjaz are cut no news. Other towns to rise on
Saturday.’ It was left to Sykes to comment upon this event. Just like the
Foreign Office officials, he was rather wary. True,
if Husayn’s revolt was successful, it constituted a serious threat to
the central powers but, if it was not, then the repercussions for Britain would
be very serious indeed. It seemed to him that, now that Husayn had
burned his boats, Britain had no other option but to support him, and ‘failure
to support the movement adequately will be disastrous to our prestige such as
it is and react to our permanent detriment in India, Egypt, and the Persian
Gulf’.⁷⁹ That the chances of Husayn failing were very real became
clear from a further telegram from McMahon. He informed the Foreign Office of
his fear that ‘both in organization and armament of forces too much
has been left to the last moment and to luck’. Although Clerk considered this
telegram ‘not altogether reassuring’, he nevertheless spotted a ray of hope,
‘we may remember that the Turks have often found it a heavy task to quell the
Arabs when circumstances were much easier for the Porte’.⁸⁰
Nobody in the Foreign
Office commented on the fact that Hussein had started his revolt without
awaiting the results of the negotiations between Great Britain, France, and
Russia, on the basis of which, as Nicolson had written to Hardinge on
16 February 1916, ‘we shall really have to come to some decision as to further
conversations with the Grand Shereef’.⁸¹ The Emir of Mecca as yet was unaware
of the agreement that, according to Grey and his officials, had to be concluded
before the Arabs could again be approached on the conditions under which they
would be prepared to revolt against their Turkish overlords.
The Arab Bureau
In the course of his
mission to the Middle East, Sykes became convinced that, if Britain did not
want to lose the war, then it was of vital importance that the activities of
the various agencies involved in the Middle East be coordinated. As he
formulated it in a speech to the House of Commons after his return, ‘if we
muddle, if we go on muddling, and if we are content to allow muddling, it will
not be a question of a draw, but the War will be lost’.⁸² After his visit to
Aden in July 1915, Sykes reported to Caldwell that he was greatly worried about
‘the want of co-ordination in our Arabian policy’. He was ‘well aware of the
departmental difficulties which lie in the way, but at the present moment the
necessity of co-ordination is of great importance especially because our enemy
is working […] from one center of political and military
influence’.⁸³ He also voiced his worries in a private letter to Clerk:
We live in watertight
compartments, if we are to do any good we should have a special committee with
one F.O. and one Government of India Representative under a person of grasp,
that is a committee of 3, established in Cairo charged with running anti- Jehad
policy – working all the Anglophil influences and anti-C.U.P. and
revolutionary influences and formulating a policy and working it, not merely
reporting […] there must be organization and coordination.⁸⁴
On his way back from
Mesopotamia, Sykes wrote a long private letter on the subject to Lord Robert
Cecil, the parliamentary undersecretary of state for foreign affairs. He
explained that:
The thing which
remains first and last in my mind, in fact, is ever present, is the want of
co-ordination which runs through the whole of
our organization between the Balkans and Basra – which is opposed to
the German scheme of things which is highly coordinated, though evidently
well decentralized. Thus in Afghanistan, Persia, Mesopotamia, Southern
Arabia, the Balkans, and Egypt you find the [undecipherable; ] committee
machine working armies, agents, and policies with one definite purpose in
accordance with a general plan, our opposition to this consists of different
parties putting up a local offensive or defensive on almost independent lines,
and quite oblivious of what the others are doing. This let me say is no fault
of any individual but the result of our traditional way of letting various
officers run their own shows, which was all right in the past when each sector
dealt with varying problems which were not related, but is bad now that each
sector is dealing in reality with a common enemy.
To counteract these
centrifugal forces, Sir Mark suggested that:
It would be worth
considering whether a new department under a secretary or under-secretary of
state should not be started, this would be the department of the Near East and
would be responsible for policy and administration of Egypt, Arabia, and
Mesopotamia […] You will notice that the area I suggest is one in language and
practically in race and its unification under one department would give the
government of the day an engine to deal with the Arab situation both national,
strategic, and economic, a personelle of side and intimate
acquaintance with the problems, and consequently give English statesmen an
opportunity of following a consistent line.⁸⁵
Sykes subsequently
unfolded a much less ambitious scheme in a telegram to Callwell of 9
October. He reiterated that he was impressed by the necessity:
For the coordination
of our policy in regard to the Ottoman Empire, Arabian people and the Mahometan
opinion in the British Empire. A means of ameliorating the position which
suggests itself to me would be to authorise me […] to complete my mission
by establishing in Cairo a Bureau under your department which should receive
copies of all telegrams giving available information regarding our enemies,
Islamic propaganda and methods and effect thereof, as well as tendency of
popular opinion, from intelligence and political officers in Mesopotamia, and
Persian Gulf, Indian Criminal Investigation Department, Soudan Intelligence
Department, Chief Intelligence Officer, Mediterranean Expeditionary Force,
Intelligence Officer, Athens. I could then from time to time transmit to you
for the use of the Cabinet a general appreciation. I suggest Egypt as the place
for the Bureau owing to its central situation and the local touch with the
Islamic world.
The Foreign Office
received both Sykes’s private letter to Clerk and his telegram
to Callwell on 12 October. Clerk considered Sir Mark’s scheme of a
special committee under ‘a person of grasp’ a good suggestion, and Nicolson
concurred – ‘the Bureau seems to me a good idea,' but they were of the opinion
that this project could not usefully be discussed before Sykes returned to
London.
For the moment, Clerk
did no more than inform Hirtzel of Sykes’s idea.⁸⁶
Sykes also raised the
subject of want of coordination in his statement before the War Committee on 16
December (see section ‘Enter Sir Mark Sykes’, above). Twelve days later, he
informed Clayton on the way things were going with, as he now called it, the
‘Arab Bureau’. It seemed that:
We are confronted with
a difficult problem. The W.O., F.O. and I.O. are slow and the Admiralty has
barged in and seized me and the Bureau […] The Admiralty want to annex the
Bureau as part of their immense network and keep me in an office in London,
they object to my organization and say all that must be left to you,
this is merely a cliché, but they refuse to transmit any suggestion of mine to
you. The objection to the Admiralty is that it is discredited, with the more
staid departments, and cannot carry the day where policy is involved. The merit
of the Admiralty is that it alone achieves anything, has large funds and does
things. Fitzgerald is of opinion that the Bureau should be nominally under the
F.O. but in fact in close touch with the D.I.D. and able to use its codes,
agents, and machinery. I have there- fore to try and pull this off but the
difficulties are immense.⁸⁷
Sir Mark’s proposals
on the constitution and functions of the Arab Bureau were discussed at an
interdepartmental conference on 7 January 1916, with representatives of the
Foreign Office, the War Office, the India Office and the Admiralty, as well as
Sykes, being present. The participants agreed that the establishment of an Arab
Bureau was desirable, and concurred with Sykes’s suggestions concerning its
functions:
The first function of
the Bureau will be to harmonize British political activity in the
Near East, and to keep the Foreign Office, the India Office, the Committee of
Imperial Defense, the War Office, the Admiralty, and the Government of India
simultaneously informed of the general tendency of Germano-Turkish policy.
The second function
will be to coordinate propaganda in favor of Great Britain and the
Entente among non-Indian Muslims without clashing with the susceptibilities of
Indian Muslims and the Entente Powers.
The bureau, however,
should not, as opposed to what Sykes had suggested, become a new, independent
agency ‘nominally under the F.O.'. The DID had its way. It was decided that the
Bureau ‘should be organized as a section of the existing Soudan Intelligence
Department in Cairo, and that it should make its reports through the High
Commissioner of Egypt to the Foreign Office’.⁸⁸ This implied that ‘Mark Sykes
drops out’, as Hirtzel wrote to Grant of the Indian Foreign
Department the same day.⁸⁹ This was quite a relief to Hardinge. He
confessed to Nicolson that he had at first been opposed to the whole scheme,
‘because we considered that the composition was radically bad, for we have no
faith in Mark Sykes […] Now that Colonel Clayton is to be head of the bureau,
we accept the position gladly, and we intend to depute a really first class
officer to represent Indian views.’ ⁹⁰
As far as the
Government of India was concerned, matters were still not settled. They were
anxious lest the first function of the Arab Bureau – to harmonize political
activity in the Near East – gave the bureau too wide powers. The bureau’s
functions should be confined to the collection and distribution of information
for the benefit of the relevant departments, and a shared responsibility about
propaganda. The viceroy, therefore, asked the India Office for assurance that
‘our political officers will not be called upon to act at dictation of bureau
without consulting Government of India’. The India Office had no difficulty in
giving it, and the Foreign Office concurred with the India Office’s suggested
reply.⁹¹
The British Indian Expeditionary Force D
In the next three
months, the Government of India succeeded in further and further reducing the
Arab Bureau’s possible influence on developments in Mesopotamia. First, the
India Office and the Foreign Office decided, at the suggestion of Sir Percy
Lake, that with respect to ‘Arab propaganda in the East’, the Arab Bureau
should do no more than to lay down principles, and that the GOC-in-C, Force ‘D’
‘should be left to make his own arrangements as a matter of local detail’,⁹²
while shortly afterwards, this time at the request of Sir Percy Cox, the
Foreign Office and the India Office agreed that liaison officers sent to
Mesopotamia by the Arab Bureau should not directly report to Cairo, but through
Cox and Lake. As Hardinge explained to Chamberlain, it was only in
this manner that co-ordination between the Arab Bureau and the political
officers, Force ‘D,' could be secured.⁹³
By the end of May
1916, the Arab Bureau, instead of the ‘committee of 3, established in Cairo
charged with formulating a policy and working it’ that Sykes had envisaged in
September 1915, had pretty much turned into one of those ‘watertight
compartments’ charged with ‘merely reporting’ that Sir Mark, using his
proposals, had wanted to abolish.⁹⁴
1. Chamberlain
to Hardings, private,25 November 1915, Hardinge Papers, Vol.121
2. Clerk’s minutes of
meeting Nicholson committee with Georges-Picot, on 23 November 1915, and
minute Nicholsen, 27 November 1915, FO 371/2486/181716.
3. Te.McMahon to
Grey, no 736,30 November 1915, and minutes Clerk 1 December 1915,
and Hirtzel, not dated, FO 371/2486/181834.
4. Nicholsen to
Clerk, 10 December 1915, and te. Grey to McMahon, no.961, 10 December
1915, ibid, Without any explanation, Hirtzel’s modified proviso was
not incorporated into the Foreign Office telegram. A copy of Sir Henry’s letter
to Hussein arrived in London at the end of December. Chamberlain protested
against the passage that was based on point 5 of McMahon’s telegram of 30
November. He wrote to Grey that:
Are we not getting
into a great mess with these negotiations of McMahon? He has now informed the
Grand Sheriff that ‘you may rest assured that Great Britain has no intention of
concluding any peace in terms of which the freedom of the Arab peoples from Germany
and ‘Turkish’ domination does not form an ‘essential condition’. Has he any
authority for this pledge?
Chamberlain had
apparently forgotten that Hirtzel had agreed to point 5, and that he
himself had approved the text of Foreign Office telegram of 10
December 1915 (underlining in original), Grey Papers, FO 800/98.
5. McMahon
to Hardinge, private, 4 December 1915, Hardinge Papers, vol. 94.
6. McMahon to
Wingate, 8 December 1915, Wingate Papers, box 135/7.
7. Clayton, Note ‘C’,
8 December 1915, FO 882/2.
8. Tel. McMahon to
Grey, no. 761, 11 December 1915, minutes Clerk, Nicolson and Crewe, 11 December
1915, and Asquith, 13 December 1915, FO 371/2486/189073.
9. Wingate to
Maxwell, private, 9 December 1915, Wingate Papers, box 135/7, Wingate to
Clayton, private, 10 December 1915, Clayton Papers, box 469/11, and
Clayton to Wingate, private, 6 December 1915, Wingate Papers, box 135/7.
10. See Roger
Adelson, Mark Sykes: Portrait of an Amateur (London, 1975: Jonathan Cape), pp.
36–9, 63–7, 95–6, 99–102, 108–16.
11. See Sykes to
Arthur, private, 12 September 1916, Kitchener Papers, PRO 30/57/91.
12. Quoted in
Adelson, Mark Sykes, p. 176.
13. Sykes to Arthur,
private, 12 September 1916, Kitchener Papers, PRO 30/57/91.
14. Report of the
Committee on Asiatic Turkey, pp. 2, 4 and 25, Cab 42/3/12.
15. M. Sykes to E.
Sykes, 3 or 9 September 1915, quoted in Adelson, Mark Sykes, p. 190.
16. Lloyd to Clayton,
private, 27 May 1916, FO 882/4, quoted in Briton Cooper Busch, Britain, India,
and the Arabs, 1914–1921 (Berkeley, 1971: University of California Press), p.
70.
17. Arnold T. Wilson,
Loyalties, Mesopotamia 1914– 1917 (London, 1930: Oxford University Press), p.
152.
18. Sykes
to Callwell, no. 19, in tel. McMahon to Grey, no. 707, 20 November 1915,
FO 371/2486/175418.
19. Sykes
to Callwell, no. 18, in tel. McMahon to Grey, no. 706, 19 November 1915,
FO 371/2486/ 174633.
20. Evidence of
Lieutenant-Colonel Sir Mark Sykes, Bart., M.P., on the Arab Question, Cab
42/6/10.
21. Secretary’s Notes
of a Meeting of the War Committee, 16 December 1915, Cab 42/6/9.
22. Crewe to Bertie,
private, 17 December 1915, and Bertie to Crewe, 21 December 1915 (italics in
original), Cab 42/6/11.
23. See: Secretary
Notes of a Meeting of the War Committee, 28 December 1915, Cab 42/6/14.
24. Lord Hankey, The
Supreme Command, Vol. II (London, 1961: Allen and Unwin), p. 469.
25. Results of the
third meeting of the Committee to discuss the Syrian question, 21 December
1915, FO 882/16.
26. Foreign Office
Note, not dated, FO 371/2486/196223.
27. Nicolson
to Hardinge, private, 30 December 1915, Nicolson Papers, FO 800/380.
28. Sykes to Clayton,
28 December 1915, FO 882/2.
29. Sykes and
Georges-Picot, Memorandum, not dated, and Nicolson, covering letter, 5 January
1916, FO 371/2767/2522.
30. Hirtzel,
Note, 10 January 1916, encl. in Holderness to Nicolson, 13 January 1916, FO
371/2767/8117.
31. Macdonogh to
Nicolson, 6 January 1915, FO 371/2767/3851.
32. Hall, Memorandum
on the Proposed Agreement with the French, not dated, encl. in Hall to
Nicolson, 12 January 1916, FO 371/2767/8117.
33. Macdonogh to
Nicolson, 6 January 1915, FO 371/ 2767/3851. In a letter to Hardinge,
Nicolson also doubted whether the negotiations ‘will ever fructify into
anything really definite’. He was personally convinced that Britain could not
‘possibly expect the Arabs to come over to our side unless we are in a position
to furnish a considerable British force to give them some stiffening’. Without
a British military intervention there was no ground for these negotiations, but
such intervention was out of the question in view of the War Committee’s
decision of 28 December. Why then continue these negotiations? Why this
dividing of the bear’s skin before it had been killed? Indeed, why this
dividing of the bear’s skin when proponents of the scheme were convinced that
the killing would never take place? In the relevant papers I have not come
across a clear-cut answer to these questions. Proponents of the scheme might
have argued that in view of French susceptibilities it was necessary to
reassure them as to British intentions, and that there was no harm in this
exercise of dividing the bear’s skin, precisely because it was highly unlikely
that the beast would ever be killed. This is also the explanation Curzon came
up with during a meeting of the Eastern Committee at the beginning of December
1918: When the Sykes–Picot Agreement was drawn up it was, no doubt, intended by
its authors […] as a sort of fancy sketch to suit a situation that had not then
arisen, and which it was thought extremely unlikely would ever arise; and that,
I suppose, must be the principal explanation of the gross ignorance with which
the boundary lines in that Agreement were drawn. Nicolson to Hardinge,
private, 16 February 1916, Nicolson Papers, FO 800/381, minutes Eastern
Committee, 5 December 1918, Cab 27/24.
34. Sykes to
Nicolson, 16 January 1916, FO 371/2767/11844.
35. Negotiations with
the Arabs, 21 January 1916, and draft agreement, not dated, Holderness to
Nicolson, 23 January 1916, Macdonogh to Nicolson, 24 January 1915 and
Hall to Nicolson 23 January 1915, FO 371/2767/14106.
36. Nicolson to Grey,
2 February 1916, FO 371/2767/23579.
37. Secretary’s Notes
of a Meeting of the War Committee, 3 February 1916, Cab 42/8/1.
38. Note, Arab
question, 4 February 1916, and Nicolson to Grey, 4 February 1916, FO
371/2767/26444.
39. Nicolson to Grey,
9 February 1916, FO 371/2767/28234.
40. Tel. McMahon to
Grey, no. 70, 26 January 1916, FO 371/2771/16451, quoted in Elie Kedourie,
In the Anglo–Arab Labyrinth (Cambridge, 1976: Cambridge University Press), p.
119.
41. McMahon to Grey,
no. 16, 24 January 1916, minutes Oliphant, 4 February 1916, and Nicolson, 5
February 1916, FO 371/2767/20954.
42. Clayton to
Wingate, private, 14 January 1916, and Clayton to Wingate, private, 17 January
1916, Wingate Papers, box 136/1.
43. Wingate to
Clayton, private, 20 January 1916, Clayton Papers, box 470/1
44. Clayton to
Wingate, private, 28 January 1916, Wingate Papers, box 136/1.
45. Minute Grey, not dated,
on McMahon to Grey, no. 16, 24 January 1916, FO 371/2767/20954.
46. I.O. to F.O., no.
P. 621, 28 February 1916, minute Grey, not dated, FO 371/2767/39490.
47. Nicolson to Grey,
2 March 1916, FO 371/2767/40645.
48. Grey to Buchanan,
no. 36, 23 February 1916, FO 371/2767/35529.
49.Tel. Buchanan to
Grey, no. 345, 10 March 1916, FO 371/ 2767/47088.
50.Tel. Buchanan to
Grey, no. 351, 12 March 1916, FO 371/2767/47950.
51. See tel. Buchanan
to Grey, no. 382, 17 March 1916, FO 371/2767/51736.
52. Secretary’s Notes
of a Meeting of the War Committee, 23 March 1916, Cab 42/11/9.
53. Draft Conclusions
of a Meeting of the War Committee, 23 March 1916, FO 371/2768/57783.
54. Tel. Buchanan to
Grey, no. 435, 27 March 1916, FO 371/2768/58401.
55. Tel. Buchanan to
Grey, no. 471, 3 April 1916, FO 371/2768/63342.
56. Tel. Buchanan to
Grey, no. 355, 13 March 1916, minute Oliphant, 14 March 1916, and tel. Grey to
McMahon, 15 March 1916, FO 371/2767/48683.
57. Tel. McMahon to
Grey, no. 204, 21 March 1916, minute Olpihant, 22 March 1916, minute Grey,
not dated, and tel. Grey to McMahon, no. 215, 22 March 1916, FO 371/2767/54229.
58. I.O. to F.O., no.
1076b, 28 March 1916, minutes Oliphant and Nicolson, 29 March 1916, FO 371/
2768/59268.
59. Tel.
G.O.C.-in-C., Force ‘D’ to S.S.I., 1404 B., 30 March 1916, encl. in I.O. to
F.O., no. P. 1181, 31 March 1916, FO 371/2768/61639.
60. Tel. McMahon to
Grey, no. 232, 1 April 1916, memorandum Chamberlain, 3 April 1916 (underlining
in original), Grey to Nicolson, not dated, and tel. Grey to McMahon, no. 263, 5
April 1916, FO 371/2768/62377.
61. Tel. Buchanan to
Grey, no. 377, 16 March 1916, FO 371/2767/51288.
62. Tel. Grey to
McMahon, no. 287, 14 April 1916, FO 371/2768/70889.
63. Tel. McMahon to
Grey, no. 278, 20 April 1916, FO 371/2768/76013.
64. Tel. McMahon to
Grey, no. 284, 22 April 1916, minutes Oliphant, 23 April 1916 (underlining in
original), Nicolson and Crewe, 24 April 1916, and tel. Grey to McMahon, no.
339, 27 April 1917, FO 371/2768/76954.
65. Tels McMahon
to Grey, no. 329, 4 May 1916, and Grey to McMahon, no. 371, 6 May 1916, FO 371/
2768/84855.
66. Minute Oliphant,
23 April 1916 (underlining in 66 original), FO 371/2768/76954.
67. Minutes Clerk and
Nicolson, 3 May 1916, and Kitchener, not dated, on McMahon to Grey, no. 83, 19
April 1916, FO 371/2768/80305.
68. Memorandum Sykes,
8 May 1916, FO 371/2768/87247.
69. Tel. Grey to
Bertie, no. 350, 11 May 1916, FO 371/2768/92354.
70. Grey to Cambon
and Cambon to Grey, 15 May 1916, and Grey to Cambon, 16 May 1916, E.L. Woodward
and R. Butler (eds), Documents on British Foreign Policy 1919–1939 (DBFP),
First Series, Vol. IV (London, 1952: H.M. Stationary Office), pp. 244–7.
71. Minute Clerk, 15 May
1916, on McMahon to Grey, no. 86, 25 April 1916, FO 371/2768/87999.
72. Tel. McMahon to
Grey, no. 414, 30 May 1916, FO 371/2768/103983.
73. Report Hogarth,
10 June 1916, FO 141/461, file 1198.
74. Minutes Oliphant
and Nicolson, 8 March 1916, Chamberlain to Crewe, 8 March 1916 (underlining in
original), minutes Kitchener and Nicolson, 9 March 1916, and tel. Grey to
McMahon, no. 173, 9 March 1916, FO 371/2773/44538.
75. Wingate to
McMahon, private, 17 March 1916, Clayton Papers, box 470/1.
76. Tel. McMahon to
Grey, no. 202, 21 March 1916, minute Olpihant, 22 March 1916, and
initials O’Beirne, Nicolson and Grey, FO 371/2767/54177.
77. Tel. McMahon to
Grey, no. 388, 24 May 1916, and minute Clerk, 24 May 1916 (underlining in
original), and tel. Grey to MacMahon, no. 426, FO 371/2773/99316.
78. Tels McMahon
to Grey, no. 402, 28 May 1916 and Grey to McMahon, no. 431, 30 May 1916, FO
371/2773/102192.
79. Tel. McMahon to
Grey, no. 436, 8 June 1916, minute Sykes, 9 June 1916, FO 371/2773/111398.
80. Tel. McMahon to
Grey, no. 443, 11 June 1916, and minute Clerk, 12 June 1916, FO 371/2773/
112684.
81. Nicolson
to Hardinge, private, 16 February 1916, Nicolson Papers, FO 800/381.
82. Quoted in
Adelson, Mark Sykes, p. 203.
83. Sykes to DMO, 23
July 1915, FO 371/2486/114293.
84. Sykes to Clerk,
private, 4 September 1915 (underlining in original), FO 371/2491/148549.
85. Sykes to Cecil,
private, 4 October 1915, Sykes Papers, box 1.
86. Tel. Sykes to
DMO, no. 12, 9 October 1915, and minutes Clerk, 12 October 1915 and Nicolson,
14 October 1915, FO 371/2491/148549.
87. Sykes to Clayton,
28 December 1915, FO 882/2.
88. Establishment of
an Arab Bureau at Cairo, 7 January 1916, Cab 42/7/4.
89. Hirtzel to
Grant, private, 7 January 1916, quoted in Busch, Britain, India, p. 101.
90. Hardinge to
Nicolson, private, 18 February 1916, Nicolson Papers, FO 800/381.
91. Tel. viceroy to
S.S.I., 15 February 1916, encl. in F.O. to I.O., no. P. 570, 24 February 1916,
and F.O. to I.O., no. 36955/16, 28 February 1916, FO original), and tel. Grey
to MacMahon, no. 426, FO 371/2773/99316.
92. I.O. to F.O., no.
P. 1342, 12 April 1916, FO 371/ 2771/70419.
93. See tel. viceroy
to S.S.I., 15 May 1916, encl. in I.O. to F.O., no. P. 1835, and F.O. to I.O.,
no. 94961, 26 May 1916, FO 371//2771/94961.
94. On 22 May 1916, Sykes was attached to the
secretariat of the Committee of Imperial Defence ‘with instructions
to make a special study of the coordination of Allied political policy in the
Middle East’. This appointment was the result of informal conversations on the
problem that there was ‘diffusion of control and that cooperation is hampered
by the want of a coordinating machine which would bring those engaged in the
problem into touch with one another’. C.I.D. to F.O., 22 May 1916, FO 371/2777/97824,
and Hankey to Grey, C.I.D. 444, 5 May 1916, FO 371/2777/85174.
For updates click hompage here