Development

At the time the First World War broke out, the Ottoman government had contracted stamp production out to Bradbury Wilkinson & Company, a prominent British security printing operation. The firm designed and printed a very tasteful pictorial issue which was first issued by the Ottoman postal administration on 14 January 1914. The stamps depicted various scenes of Constantinople and its environs, excepting the 2 piastres which showed the Ottoman warship Hamidiye¹ and the 200 piastres on which appeared the visage of the Sultan himself. Once the entire order had been printed, the printing plates and etc. were placed in boxes which were locked and sealed on 26 May 1914 in the presence of a representative of the Ottoman government. ² The boxes were stored on Bradbury Wilkinson’s premises. Opening the boxes required the seals to be broken, and two locks to be unlocked — the key to one was held by Bradbury Wilkinson, the other by the Ottoman postal authorities. ³ Bradbury Wilkinson was not caught in the middle of performing any printing for the Ottoman posts when war between Britain and the Ottoman Empire broke out in November.

Bradbury Wilkinson’s designs are unarguably excellent: the quality of the central vignettes is unimpeachable and, while the tastefulness of the frames varies a little between individual designs, there's not a one of these stamps isn't overall extremely charming. The engraving and printing are both superlative, although the registration of the bi-colours varies a little in accuracy and the higher values are regularly found below-averagely centred. The entire issue is engraved, except for the lowest-value 2, 4, 5 and 6 paras which are lithographed. The few vertical designs give a slightly inharmonious appearance to the overall ensemble, and their disappearance from the occupation issue —a disappearance not prompted by aesthetic considerations, as will be seen— I think was no great loss, as nice as each of the stamps individually is. The effigy of the Sultan is really good portrait work.

We skip forward to late 1917: Baghdad was in British hands and Sir Percy Cox (Chief Political Officer, Indian Expeditionary Force and later Civil Commissioner for Iraq) had successfully brought the Baghdad issue to fruition on 1 September. Exhilarated by his philatelic triumph, he was soon dreaming of a far more ambitious sequel. He had seemingly brought a Gibbons catalogue with him on campaign and, looking through it, he noticed the fact that the 1914 Ottoman definitives had been printed in Britain. As early as 10 September 1917 he cabled the India Office:⁴

He repeated his proposal to the India Office a week later, adding by way of additional argument his belief that (with reference to the Indian stamps surcharged “I.E.F.” which were coming into use now that stocks of the Baghdad issue were running out) “’Expeditionary Force’ stamps are not really appropriate for use of civil population Baghdad Vilayet.”⁵

The India Office was unenthusiastic about the proposal. The Under-Secretary of State for India recorded his thoughts in a minute of 23 September 1917: “I am not much enamoured of the idea of manufacturing bogus Turkish stamps for the purpose of using them in Mesopotamia with a British overmark.” The matter was referred to the Foreign Office for them to determine.⁶

The Foreign Office had no objections, replying to this effect on 2 October.⁷ The Under-Secretary of State for India took the opportunity to record his doubts on the file once again (“I feel exceedingly doubtful about this proposal, as already noted, but as the Foreign Office agrees, I suppose we must go through with it”⁸) but the issue was now cleared for takeoff.

The India Office established contact with Bradbury Wilkinson on 8 October 1917, asking whether they would be in a position to produce a reprint of the 1914 issue and then overprint it.⁹ They quickly replied in the affirmative, but they noted the practical difficulty of getting at the plates now that they had been locked up in the manner described above.¹⁰ They were willing to force the locks but, anxious of offending the Ottoman government by so doing, and damaging their prospects of obtaining post-War contracts as a result, they requested an express authorisation from the British government before proceeding.¹¹ This was duly provided, and the locks were broken on 31 January 1918, in the presence of a director of Bradbury Wilkinson and representative of the India Office.¹²

The question naturally arose of what values the overprinted stamps were to be given. The 1914 Ottoman issue comprised 17 denominations, and Cox erroneously believed the Indian issue then in use also comprised 17 denominations. He therefore suggesed a simple transposition of each of the Indian values in sequence, and added “the relative Indian and Turkish face values will not exactly correspond but that in any case is not possible and as not important.”¹³ When it was pointed out to him that the contemporary Indian issue only comprised 16 denominations, he proposed leaving out the 2 paras stamp (the lowest denomination) and overprinting the other 16 values starting with the 4 paras.¹⁴ This produced the following scheme:

The India Office communicated this to Bradbury Wilkinson on 8 January 1918.¹⁵ Bradbury Wilkinson replied the next day and confirmed the instructions, setting out a table of each Ottoman denomination next to the Indian value it was to receive. It seems to have been at this point the India Office realised that the Ottoman 200 piastres was potentially a delicate matter — on Bradbury Wilkinson’s letter, next to the mention of the 200 piastres stamp, someone has noted in pencil “Sultan’s head”.¹⁶ The India Office cabled Cox asking for his thoughts, copying in the Viceroy of India, the Foreign Office and the Directorate of Military Intelligence.¹⁷ Cox’s preference was to overprint the stamp regardless:¹⁸

The Foreign Secretary eventually intervened on 1 February 1918 and forbade the use of the 200 piastres on political grounds. The Foreign Office also sensibly noted that the high value of the stamp made it “not really essential for revenue purposes”. ¹⁹ A new difficulty appeared a few days later. It was noticed at the India Office that the stamps, as overprinted, wouldn’t be compliant with the UPU colour scheme. The India Office raised the matter with the GPO, and the GPO insisted on changes being made to the scheme of overprinting so that the Indian values fell (where needed) on stamps of the correct colours. ²⁰ Essentially the “key” denominations from this perspective were the 10 paras, 20 paras and 1 piastre, coloured green, red and blue respectively — these being the UPU regulation colours for international printed matter, postcards and letters. These denominations needed to be overprinted as ½ anna, 1 anna and 2½ annas to match the Indian colour/value concordance, and the remainder of the Indian values then had to be shuffled into place around these three. The scheme was accordingly revised as follows — arriving at the scheme actually used on the issue:

The India Office cabled Cox on 5 February 1918 informing him of these proposed changes. Cox noted the revised scheme lacked a 2 annas, which in his view was undesirable in view of that value being the Indian registration fee. He requested that, if this scheme was adopted, an extra quantity of 1 anna stamps be printed. ²¹ This was agreed to.

There was a separate question, running in parallel to the above, as to the colour of the overprint: two of the Ottoman values (the 4 and 6 paras) were in dark colours and consequently a black overprint failed to show up clearly. Bradbury Wilkinson tried to see if a red overprint would work better, but it was also unsatisfactory visually, and they noted to the India Office their further concern that “red ink is difficult to get in these times”. ²² In the event this turned out to be a moot point as neither the 4 or 6 paras stamps were used for overprinting.

On 14 February 1918 the India Office instructed Bradbury Wilkinson of the revised scheme of overprinting, as set out above. ²³ When Bradbury Wilkinson actually got to work is unclear, but on 18 March they indicated they hoped to have a first shipment despatched on 25 March, and the entire order of stamps completed and sent out by 8 April. ²⁴ In the event (to heavily summarise a matter which generated considerable correspondence) these deadlines were missed, and Bradbury Wilkinson sent the stamps in piecemeal parcels of only a few denominations each, in contravention of an instruction from Cox that each parcel contain an amount of each denomination. All of the order was eventually despatched by 4 June. ²⁵

Issue and subsequent events

The stamps went on sale on 1 September 1918. ²⁶ Major Clerici (in charge of the civil postal service in Iraq — see much further down the page) maintained a sensible disdain for collectors and speculators: he ordered postmasters to refuse to sell entire sheets to customers, even those whom they knew had genuine need for large amounts of postage. ²⁷ Apart from a few specimens diverted for the Royal Collection, Imperial War Museum, UPU et al . the entire printing was sent from England to Baghdad, and stamps had to be re-imported from Iraq by dealers.²⁸

Cox ordered more stamps as early as 12 September 1918, ²⁹ and by 9 December a top-up of the ½ anna was “very urgently required”. ³⁰ On 30 December the Civil Commissioner’s office placed a monthly standing order with Bradbury Wilkinson, stating that “increased supplies proposed are required on account of reintroduction of civil parcel post, ³¹ opening of new post and telegraph offices and general expansion of business.” The Civil Commissioner requested an increased standing order on 26 August 1919. ³² This standing order elapsed with a final shipment on 14 September 1920. ³³ The Civil Commissioner estimated that these stamps should be enough to last the remainder of the financial year (i.e. to 31 March 1921). ³⁴

Cox’s letter of 26 August 1919 also requested a new “ON STATE SERVICE” overprint. The first design (of which I can find no image) was rejected on grounds of smallness,³⁵ but the second design was accepted and the stamps first arrived in the shipment of 26 January 1920.³⁶ Gibbons gives the service stamps an issue date of 16 May 1920.³⁷

Bradbury Wilkinson’s actions in breaking open the locked boxes containing the plates came back to haunt them. On 13 August 1919 they were sent an anxious letter from Mr Yaremdji, their agent in Constantinople, reporting that the Ottoman government had found out what they had done, and was “much offended” by it. The Ottoman postal ministry “[intended] to claim damages from the firm for the prejudice which has been caused to them”, and naturally the prospect of them obtaining any more Ottoman contracts was placed in very considerable jeopardy. Yaremdji requested written proof that the reprint was done solely on the orders of the British government, which he hoped would smooth matter over with the Ottomans.³⁸

Proof was duly assembled, and on 3 October 1919 the British High Commissioner in Constantinople wrote to the Ottoman Foreign Minister, informing him that “the action taken by the firm in regard to the seals and locks in question was taken by express order of His Majesty’s Government to enable them to print further stamps to be surcharged for use in Mesopotamia. Messrs Bradbury Wilkinson and Company Limited had no option but to comply with this order, and no responsibility attaches to them in the matter.”³⁹ Rather surprisingly, this was apparently enough to resolve the matter, as Bradbury Wilkinson received the contract to print what would be the final Ottoman definitive issue in 1919.⁴⁰

On 1 October 1919 the Civil Commissioner’s office requested that the ¼ anna stamp be discontinued and the 5 paras stamp instead be overprinted 1½ annas.⁴¹ These first arrived in the shipment of 5 May 1920⁴² and the exact issue date is apparently not known. Why this need arose at this point I am unsure: the basic rate for a letter to within the British Empire had been 1½ annas since 1 September 1918⁴³ and (if Proud is correct) there were no further rate changes until 1921.

The use of overprinted Ottoman designs in the post-armistice period appears to have started raising eyebrows quite quickly — as early as May 1919 (that is, over a year before the actual peace treaty) the Civil Commissioner noted to the India Office that continued use of the stamps stamps was “[giving] rise to much misapprehension”.⁴⁴ Some tentative steps towards designing a new issue were begun in mid-1919, but the matter drifted while peace negotiations continued and, without peace having been concluded, the question of what currency would be used in Iraq remaining unresolved.⁴⁵ The “1919” designs were eventually abandoned and I’ll describe these in their proper place elsewhere.

What happened after this point becomes a little obscure as the India Office files, on which I have been enormously relying, run out around here. As noted above, the stamps printed by September 1920 were expected to have run out by April 1921, and further printings were evidently ordered before the issue was finally invalidated on 1 June 1923.⁴⁶ Gibbons lists certain varieties dated to 1921, 1922 and 1923, the latest with an issue date as late as 10 March 1923.⁴⁷ I have no details beyond this.

Matters of larger importance

We'll return to the stamps shortly, as something probably needs to be said about the British occupation itself at this point. The whole Sykes-Picot thing is very well-known so I don't think too much detail needs to be gone into here, probably. The armistice with Turkey saw Britain in control of the Basra, Baghdad and (of which much more on the next page) Mosul vilayets. There they stayed while peace negotiations with the Ottoman government began.

Initially the inhabitants of the Middle East did not necessarily view the British and French as unwelcome guests, seeing as they had liberated the place from Turkish rule, but they hoped that, once the war was over, they would clear out and leave the place to its own devices. Of course, this was a rather optimistic assessment of the situation. Britain was awarded a League of Nations mandate to govern Iraq in April 1920, and when news of this reached Iraq the following month, anti-British manifestations began almost immediately. The mandate was seen as offensive on several fronts at once: the post-war division and military occupation of the Middle East by the European powers was intolerable to Arab nationalists (and to Iraqi and Kurdish nationalists, such as they were at that point), and the concept of a great swath of the Islamic world falling under the control of Christian countries was intolerable to many devout Muslims. Resistance to the mandate was initially peaceful, but the situation descended into violent revolt in June (Proud notes, rather coolly, that following a fiery anti-British speech in Baghdad "a machine gun had to be turned on the crowd" ⁴⁸ ). The revolt had a fairly ecumenical character, with Sunnis and Shias putting aside their differences to fight the British, and the rebels initially seemed to have the upper hand. The British however counter-attacked in strength (using the then-innovative strategy of employing aircraft as the main combat arm, rather than ground troops) and the rebels were defeated by October 1920.⁴⁹

Although the revolt had completely failed, it had nevertheless convinced the British that they needed to adopt a different approach to administering Iraq. Britain resolved to find a form of government which would allow it to still control Iraq, but which would be of a more palatable character to Iraqis than a military occupation. The centrepiece of the British plan was Faisal ibn Husayn (who we'll meet in more detail in page VI), a very prominent Arab military/political/religious figure who (a), had a history of co-operating with the British, and (b), was presently languishing in unemployed exile in England, his brief tenure as King of Syria in mid-1920 having been put to a violent end by the French. The Iraqi people were presented with what Wikipedia discreetly calls a "carefully arranged" plebiscite, asking whether Faisal should be made king of Iraq. A neat 96% of respondents answered yes, and so Faisal was shipped off to Baghdad and crowned King Faisal I on the 23rd August 1921. Faisal's installation as king was to be accompanied by an end to British military occupation and the installation of a civil government, and this was eventually confirmed by the Anglo-Iraqi Treaty of October 1922. And we'll see how this all worked out in practice on the next page.

Varieties and etc.

The issue was printed in a somewhat intermittent manner over a few years and, not surprisingly, some varieties are known. I’d like to do a closer look at these someday but at time of writing I lack the material. Aside from the really exotic expensive stuff (inverted centres, inverted overprints, an inexplicable red overprint etc.) a lot of the postage values can be found with overprints of two or three different widths. These varieties don’t occur in the service issues.⁵⁰

The stamps were initially unwatermarked but 1921-23 printings appeared for some reason on paper with a “Crown Agents” watermark. Some official stamps were printed in gently different shades to the corresponding postage stamps.

There is also, of course, the “wrong tughra” error, arising out of Bradbury Wilkinson having managed to maintain their contract for printing stamps for the Ottoman government. At one point in 1921 the printing plate for the new Ottoman 10 piastres was confused with the plate for the "occupation" 10 piastres, creating the error.⁵¹

Amounts printed / philatelic interest (etc.)

Since my last draft of this page I’ve revised my opinion somewhat vis-à-vis how “philatelic” this issue was. It was devised for political reasons, quite explicitly, but I count in its favour:

• The stamps clearly did do a lot of sincere postal service, that was never in doubt.
• The India Office files reveal a demand for higher values to pay telegraph fees.
• In general, the files suggest genuine demand for the stamps. No doubt some of this was driven by collector and dealer traffic, but one finds references to specific postal needs, denominations being exhausted at different speeds, etc.
• The stamps weren’t offered directly for sale by the responsible authorities in the UK.
• Contrary to the assertion of Squadron-Leader Ennis, as cited by Cockrill,⁵² the original order for the issue did not request the same number of each denomination. There were only (for example) 50,000 each of the 5 and 10 rupees ordered, compared to 600,000 of the 1 anna.⁵³

Regarding the last point: Cockrill was unable to find the actual amounts printed.⁵⁴ I believe I can improve somewhat on this: the India Office files allow an attempt at reconstructing the amounts printed up to September 1920. Of course, this is far from the entire issue, which as noted was printed into 1923. But the figures, to the limited extent they’re available, may be of some interest, so I set them out below. The “by 2 October 1919” column is taken from the materials prepared in connection with the Ottoman government complaint, and subsequent columns come from Bradbury Wilkinson delivery notifications sent to the India Office.

Now, all this having been said, the stamps are available readily enough in unused condition, although the top values are a little pricey (anecdotally the 5 rupees appears harder to find than the 10 rupees, despite Gibbons pricing it about one-third as much). Collector interest in such a handsome set is readily understandable, and Cockrill claims it was also a highly popular souvenir for soldiers posted to Iraq.⁵⁵ The 5 and 10 rupee denominations seem to have been supplied somewhat in excess of both postal needs and collector enthusiasm, regardless of how many were needed for telegram receipts — when the issue was invalidated in 1923 there was still "25 years' supply" of them sent back to Baghdad to be destroyed.⁵⁶ Today the lower values are generally cheap for both mint and used specimens. In this case I've elected to show mint examples, where I have them, owing to the extreme niceness of the designs. The highest values are more expensive unused, in the normal fashion.

Use in Mosul vilayet

Owing to the political delicacy in the Mosul region (see next page) the issue was withdrawn from there on the 1st February 1919⁵⁷ and replaced with the special Mosul issue. The Mosul stamps ran out at some point after 1919⁵⁸ and the regular occupation issue then reappeared.

Telegraph uses The stamps were used for telegraph purposes, and Cockrill notes that this was a common for the higher values. ⁵⁹ It seems to have taken some time before there were enough stamps in Iraq that a quantity could safely be diverted from the stocks needed for postal use, however. On 5 December 1919 (that is, over a year after the stamps were first issued), the Civil Commissioner complained that “present supply is inadequate. I have been unable in consequence to introduce payment for telegrams by means of stamps and have had to arrange to debit civil and military departments for telegrams by estimation of average traffic which is unsatisfactory.” ⁶⁰

Revenue uses

The issue was overprinted in a couple of formats for revenue purposes. ⁶¹

Some curiosities

Gibbons (2009) claims that the revenue stamps were pressed into postal service during a cholera outbreak in Baghdad in January 1923, while Proud (1996) explicitly denies that this happened. ⁶³ I've seen one or two covers "in trade" purporting to be from this time of cholera, bearing what appear to be genuine revenue stamps and correct postmarks, whatever that is worth. I'd be inclined anyway to assume this was a mere philatelic contrivance, if it did genuinely happen.

Gibbons also notes (but doesn't give a number to) disreputable bisects which apparently only appear on philatelic items — I've seen a few on sale at points and all are appropriately neat-looking. According to Cockrill these bisects were unauthorised but did manage to pass through the post in a few places in 1919. ⁶³

Gibbons' online auction of 29 June 2021 included a 15 January 1919 cover franked with, instead of a stamp, a "tab" from the bottom of a 2½ annas sheet, reading "2½ An" as the bottom margin did in a couple of places. This was postmarked, seemingly entirely correctly, at FPO 339 (Makina stationery office, Basra) and addressed to Ashar, Basra. Like the bisect covers mentioned above, this seems like a genuine item, albeit one obviously contrived for philatelic purposes.⁶⁴

Evolution of postal administration in Iraq

Cox gave a summation of postal administration as of 30 November 1917: in Basra vilayet there was a “well-established system of civil offices working parallel with, but independent of, field post offices”, but in Baghdad vilayet the only civil post offices (plural) were within the city limits, the other post offices in the vilayet all being field post offices. ⁶⁵ A system whereby mail could be collected from elsewhere in the Baghdad vilayet began in June 1917, but throughout 1917 there were no actual post offices outside Baghdad itself. ⁶⁶ Subordinate positions in the civil offices were staffed, as far as possible, by locals, with other roles filled by personnel seconded from Civil offices in India. ⁶⁷

Civil post offices in the Basra vilayet used unoverprinted Indian stamps. ⁶⁸ Civil post offices in the Baghdad vilayet seem to have used Indian stamps overprinted “I.E.F.” once the Baghdad issue ran out. ⁶⁹

The entire civil postal apparatus in Iraq was at this time subordinated to the Indian civil posts, which Cox regarded as increasingly unsatisfactory. In the 30 November telegram quoted above he proposed that civil posts in Iraq be detached from India and administered by their own civil director, who would report to the military commander in the theatre. This proposal was agreed and On 1 June 1918 Major Clerici was appointed Deputy Director of Postal Services (Civil), Iraq. ⁷⁰ By 28 June 1919 he had been elevated to "Director of Postal Services, Iraq and Persian Lines of Communication. ⁷¹

I find this a little hard to follow but it seems that Iraq was treated as part of India for rates purposes until 1 September 1918, whereupon it was treated as a separate part of the British Empire and the “imperial rate” applied where applicable, international rates where it didn’t. ⁷²