Taken from here: http://www.masud.co.uk/ISLAM/bmh/BMH-AQ-offa.htm
Sheikh Abdullah Quilliam, writing as Professor H. M. Léon, M.A., D.C.L., F.S.P., etc. (1916)
In the Numismatical department of the British Museum there is preserved a curious and interesting gold coin, over twelve hundred and thirty years old, on which is inscribed in unmistakable Arabic characters the declaration that ‘There is no Deity but Allah, The One, Without Equal, and Muhammad is the Apostle of Allah,’ and the further declaration, engraved around the margin of the coin, ‘Muhammad is the Apostle of Allah, Who sent him (Muhammad) with the doctrine and the true faith to prevail over every other religion.’
This coin was engraved, struck and issued by Offa, King of Mercia, or ‘Middle England’ (an ancient Anglo-Saxon kingdom, which extended on both sides of the River Trent from the North Sea to Wales), from 757 to 796. The name, originally restricted to the district around Tamworth and Lichfield and the Upper Trent valley, refers to a ‘march,’ a moorland, or frontier, which had to be defended against hostile neighbours; in this case such ‘alien enemies’ being the Welsh, the ‘Ancient Britons,’ who for centuries contended with the Anglo-Saxon invaders for supremacy in that region.
A number of smaller states were gradually incorporated with Mercia, the first settlements being probably made during the second half of the sixth century of the Christian era. The kingdom was, however, of but little importance until the accession of Penda in 626 (C.D.), who rapidly, by his vigorous policy and equitable rule, attained a supremacy over the other kingdoms, particularly after his victory at Hatfield (or Heathfield) over Edin, the powerful Deiran king, in 633. In 655, however, Penda was defeated and slain at Winwaed by Oswin, king of Northumbria,  and for the time being Mercian supremacy was terminated. Wulfhere, the nephew of Penda (659-675), pushed back the Northumbrians, and extended the boundary of the kingdom southward to the Thames. Wulfhere was the first monarch of this kingdom to renounce paganism and embrace Christianity. One of his successors, Ethelbald (716-757), further spread the boundaries of Mercia, by making large encroachments upon the territories of adjoining states. But the mightiest kings of Mercia were Offa (757-796) and Cenwulf (806-819). After the death of the latter monarch the kingdom rapidly declined, and in 828 it was merged in the realm of Egbert, king of Wessex.
King Offa, in whose reign the interesting coin we have under consideration was struck, succeeded to the throne of Mercia in 757, he being the ninth monarch of that kingdom in succession from Wybba, the father of Penda (to whom allusion has previously been made). He found the kingdom much weakened, and probably the early years of his reign were occupied by him in restoring rule and order within his territory. In 771 he began a career of conquest: he defeated the army of the King of Kent in 775, and fought successfully against the West Saxons (779) and the Welsh. As a protection against these lattter marauders he constructed a great earthwork which extended along the whole border between England and Wales, from the north coast of Flintshire, on the estuary of the Dee, through Denbigh, Montgomery, Salop, Radnor and Hereford, into Gloucestershire, where its southern termination is near the mouth of the Wye. Portions of this rampart still stand to a considerable height, though much of it has been almost obliterated by the ravages of time, the elements, and human beings. A vast amount of labour must have been expended to construct this work. Nearly parallel with it, some two miles to the eastern or English side, is an inferior rampart termed Watt’s Dyke, which was also constructed by Offa and completed about 765 (C.D.). It is conjectured that the space between the two dykes may have been a species of neutral zone for trading purposes.
Offa had cordial relations with the Roman See. Two Legates, George and Theophylact, visited Mercia, and were received by the king at a court held at Lichfield in the year 786. The report which these ecclesiastics made to Pope Adrian I, attributed to 787, is printed in Birch’s Cart. Sax., No. 250. In this document there is direct reference to the vow made by King Offa to Pope Adrian I, through the Legates to send 365 mancuses to the ‘Apostle of God’ (i.e. the Pope), ‘as many as there are days in the year, as alms for the poor, and for the manufacture of lights for the church.’
This donation by Offa appears to have been the origin of what has ever since been known as ‘Peter’s Pence,’ and won from the Pope the grant of a Mercian archbishopric.
The importance of this grant by Offa will be hereafter seen, when we come to more particularly discuss the origin of the interesting coin we have now under consideration.
The coinage of the kingdom of Mercia appears to have been the most important of all the coin-striking kingdoms of the Heptarchy. The earliest Mercian coins are those which belong to the sceatclass. These were usually of silver,  and weighed from eight to twenty grains.
These early Mercian sceattae bear the names of Penda and Ethelred. The coins of the former are of purely Roman types, but those of the latter show a mixture of Roman and native design, thus pointing to a somewhat later date. The inscriptions on Penda’s coins are in Roman and Runic characters, but those of Ethelred are in Runes (the ancient alphabet of the heathen Northmen) only.  The name of the king in each instance is given on the ‘reverse’ side of the coin.
From the death of Ethelred (704) to the reign of Offa (757-796), a period of over half a century, there are no numismatic records of Mercia.
Offa did not strike any sceattae, and his coins mainly consist of the ‘penny’ class. They were of silver and weighed from eighteen to twenty grains. It is believed that Offa was the first monarch to introduce the ‘penny’ into England. The form of this coin, but not the type, was derived from the denier of Charlemagne. 
Offa’s coins of the ‘Penny series’ are remarkable for their artistic excellence both in execution and design, and in this respect far surpass the issues of many succeeding rulers. The types are not only numerous but varied. They can be classified into two series: those bearing the bust of the king, and those in which the bust is absent. The bust, when present, is original in character, and exhibits undoubted attempts at portraiture. The designs on the reverse of the coins are decidedly ornamental, and comprise for the most part elaborately formed crosses or floral patterns. The busts upon the coins are well formed, and the head bears a life-like expression, the hair being usually arranged in close curls or plaits, but in some of the specimens it is loose and flowing. The inscriptions are generally in Roman characters, but here and there traces of Runes survive. There are no indications of mint-names, but we may conclude that the principal Mercian mint was in London. The coins themselves, however, prove that after the defeat of the King of Kent and his army in 775, at Otford (about three miles north of Sevenoaks and eight-and-a-half miles north-north-west of Tunbridge, Kent), when Kent became a fief of Mercia, Offa made use of the Canterbury mint. 
The remarkable gold coin of Offa bearing the Arabic inscription has furnished much food for reflection amongst the students of numismatics, and it is generally conceded that it is one of the rarest and most remarkable coins in the world.
Many treatises and papers have been written upon the coin and its origin, and numerous theories propounded with regard to the same.
So far back as November 25, 1841, a paper by Monsieur Adrian de Longperier, of Paris, was read before the Numismatic Society of London  upon this very coin. Mr. J. Y. Akerman, in a paper read before the same society on March 24, 1842, and printed in the Numismatic Chronicle, vol. v. pp. 122-124, also considers the raison d’être of this coin being struck. It is referred to by Mr. Herbert A. Grueber, F.S.A., in his Handbook of the Coins of Great Britain and Ireland in the British Museum (published 1899).  The coin is fully described in Kenyon’s Gold Coins of England, 1884, pp.11, 12, and is illustrated in the frontispiece to that work, Fig. 13. It was made the subject of an exhaustive and extremely interesting article by Mr. P.W.P. Carlyon-Britton, F.S.A., President of the British Numismatic Society,  and as recently as 1914 was the subject of an excellent paper by Mr. J. Allan, M.A. (British Museum Staff, Coin Department). 
The theories put forward by the above learned gentlemen, all of them well versed in numismatical lore, and by some other individuals, whose names are not so well known to fame, may be classified under the following heads:-
(1) That Offa had become a convert to Islam, and took this means of declaring his acceptance of that Faith by stamping the Kalima, or Islamic Confession of Faith, upon his coins.
(2) That, without knowing the meaning of the Arabic words upon the coin, possibly merely regarding them as so much ornamentation, Offa had the coin struck off, merely adding, in order to identify himself with the same, the words ‘Offa Rex’ stamped also thereupon.
(3) That, as many pilgrims proceeded from England to the ‘Holy Land’ of Palestine, then under the dominion of the Muslims, this coin was struck, bearing this Arabic inscription in order that it might be the more readily accepted by the Muslims, and thus facilitate the journey of the pilgrims are assist them in trading (which may of them did) in those lands.
(4) That the piece was not a coin intended for general circulation, but was struck specially as a mancus and as one of the quota of 365 gold pieces which Offa had vowed to pay annually to the Pope of Rome.
Undoubtedly something can be said in support of each of these theories, but in these matters one must carefully consider all the circumstances of the time and weigh the pros and cons upon the subject. The first theory, namely that Offa had accepted Islam, appears to me to be absolutely untenable. At that time Islam was naught more to the Western world than the absolutely living embodiment of ‘Anti-Christ.’ Its tenets were not only not understood, but were wickedly misrepresented, and it was this wilful misrepresentation of ‘The Faith Most Excellent,’ and the colossal ignorance and superstition of the mob, that made the series of crusades possible. Even to-day, twelve centuries after the passing away of Offa, the most profound ignorance exists among the masses as to Islamic doctrines and ethics. Is it then at all probable that Offa, who had petitioned the Pope to grant him an Archbishop for his kingdom, and had voluntarily vowed to pay 365 golden pieces each year to that pontiff (and we know from authentic documents and records that up to his death such tribute was regularly paid by Offa), and had received with open arms and the greatest honour the legates from Rome, should become a Muslim? At that period in the world’s history for Offa to have done so would have meant for him, not merely the loss of his throne, but probably his life also.
Most of his other coins are stamped with a cross and bear his bust! That is not very Islamic.
True that the cross may have been placed upon the coins, and deeply indented therein, so as to enable the same to have been the more easily divided into halves or quarters; but the cross is there, and we cannot conceive any ‘True-Believer’ placing such an emblem upon any coin issued by him.
Furthermore, after the conquest of Kent by Offa in 775, and the adding of that territory to his kingdom, we find the Archbishops of Canterbury acknowledging Offa (and subsequently his successor Coenwulf) as their overlord. This is amply proved by one of the coins struck by the Archbishops of Canterbury (who possessed the right of minting money) at that period.
Jaenberht (766-790) is the first Archbishop of Canterbury of whom coins are known. During his episcopate Offa conquered Kent, and as Jaenberht’s coins were struck under his supremacy, they always bear that ruler’s name on the reverse. The obverse types are a star, a cross potent or pommée, or the name of the archbishop in three lines only. The reverse is always the same with one exception, namely, with Offa’s name at the end of a cruciform object.
The next archbishop was Aethelheard (793-805); he was elected to that office in 791, but did not receive the pallium until 793. During this interval he appears to have struck coins with the title of Pontifex instead of Archiepiscopus. His early coins bear the name of Offa; but those struck after 796 that of Coenwulf. Those with the name of Offa have for obverse and reverse types:- a star, a cross, the Christian monogram, etc. 
Is it likely that these archbishops, whose territory had been conquered by Offa, who had set up a rival archbishop to them in his own dominions, would have put the name of Offa on their coins if he had accepted Islam? Rather would we not have seen them denouncing him as ‘an infidel,’ and rousing the populace in revolt against him and his rule?
The first theory therefore appears to be absolutely untenable.
Let us now consider the theory that without knowing the meaning of the Arabic words upon the coin, and possibly regarding them as pure ornamentation, Offa had the coin struck, adding the words ‘Offa Rex’ to the original superscription.
Mons. Adrian de Longperier inclines to this view. He says:-
‘However strange this piece may appear, it is yet susceptible of explanation. The faults of orthography to be traced in the legend, which is reversed in its position with the words OFFA REX, show that it is a copy of a Mussulman dinar, by a workman unacquainted with the Arabic language, and indeed ignorant of the fact of these characters belonging to any language whatsoever. Examples of a similar description of coin were put in circulation by the French bishops of Agde and Montpellier in the thirteenth century. In the present case, we cannot see an intentional adoption of a foreign language, as on the coins of Russia, Spain, Sicily, Georgia, and even Germany. On the moneys of Vassili Dimitrivitch, of Dmitri Imamvicht, on that of the Norman princes William and Roger, and the Mozarbic dinar of Alfonsus, we find Arabic legends appropriated to the very princes by whose commands they were struck. One silver piece of Henry IV, Emperor of Germany, bears on the reverse the name of the Khalif Moktader Billah-ben-Muhammad; but this is merely the result of an association between those princes.’ 
In support of the views of Mons. A. de Longperier, it is worth noticing that in later times there were issued by Christian princes coins having inscriptions partly in Roman and partly in Arabic characters, and some were issued by Crusaders with entirely Arabic inscriptions.
In Mr Carlyon-Britton’s paper upon this coin  he quotes five specimens of this description of coin, namely:-
(1) A gold coin of Alfonzo VIII, of Castile (1158-1214, Christian date), the Arabic inscription on the obverse side whereof reads thus: El Imam al- bay’ata el mesiahyata el Baba ALF. Bismiel ab Walibu wa errooh el kaddûs Allahoo wahido mam aman wa’ tamada yekdon salminan. The translation whereof if: ‘The pontiff of the church of the Messiah, the Pope. ALF. In the name of the Father, and of the Son, and of the Holy Ghost, one God. He that believeth and is baptized shall be saved.’
The reverse side of the coin bears an inscription in Arabic of which the following is a translation: ‘Prince of the Catholics (Amir el Katolikin), Alfonso, son of Sancho. May God help and protect him.’ Around the margin of the coin we find this legend also in Arabic characters: ‘This dinar was struck in the city of Toledo, 1235 of (the era of) Assafar.’
The era of Assafar dates from 30 B.C., that being the date of the submission of Spain to the Romans, consequently the coin in question dates from the year 1197 of the Christian era.
The other coins exhibited by Mr Carlyon-Britton are:
(2) Silver ‘staurat’ drachma struck at St. Jean d’Acre about 1251 under Louis IX (1251-1259).
(3) Gold besant struck by Crusaders at St. Jean d’Acre in 1251.
(4) Early imitation by Crusaders of dinars of El Amir (Fatimite Khaliph from 1101 to 1130 C.D. = 494 to 524 Hegira), and attributed to the regency of Bohemund I of Antioch under Tancred.
(5) Imitation of a dinar of about the time of Hisham II (Hegira 400-403), independent Amawu Caliph in Spain. Found in Spain.
Of the above five coins, three of them (1 to 3) contain Christian inscriptions written in Arabic language and character; the latter two (4 and 5) are written in corrupted Arabic. No.4 has a small Maltese cross in the centre of the reverse side.
The third suggestion, namely that the coin was coined by Offa for the use of such of his subjects as made the pilgrimage to the ‘Holy Land,’ does not seem very probable. The number of such pilgrims, of necessity, would be limited to a comparative few, and the monarchs of those days, even if they were as pious as King Offa is stated to have been, were not distinguished for any particular solicitude for the comfort of their subjects. The majority of these rulers would rather grind out from their subjects the uttermost farthing they could extort, rather than go to the expense and trouble of providing special coins for their use while on a pilgrimage.
There remains, therefore, but the fourth proposition to consider, and here we find ourselves on much surer ground. We have already seen that Offa had made a vow that he would pay 365 gold pieces every year to the Pope, and that probably, in consideration of the faithful fulfilment of that vow, the occupant of the pontifical throne had bestowed an Archbishop upon Mercia. The exact date of Offa’s vow we do not know, but it may fairly be presumed that he made it to the two papal legates, George and Theophylact, who visited him in 786.
The date upon Offa’s coin now becomes extremely important.
The coin bears the date Hegira 157, equivalent to 774 of the Christian era. This date does not, however, prove that Offa’s coin was struck in that year (twelve years prior to the visit of the legates); but as the piece is manifestly a copy of an Arabic dinar of that year (Hegira 157), made by a person who did not understand Arabic (otherwise, why did he place the words OFFA REX in an inverted position to the Arabic characters), all that the date, 157 Hegira, demonstrates is that Offa’s coin was struck in, or, what is more probable, subsequently to the year 774 of the Christian era.
What appears to us to be the most probable origin of this coin is that when Offa made his vow, the question arose as to what was to be the size and weight of each of the 365 ‘gold pieces.’ In reply to such a query on the part of the king, who would naturally desire to know the exact extent of his liability, what would be more natural for one of the legates to hand Offa a coin, and say, ‘365 gold pieces like this’?
Arabic coins were well known at Rome. Countless pilgrims from the Holy Land passed through the Eternal City on their return from Palestine, many of whom laid offerings at the foot of the papal throne. It is fair to presume that amongst such offerings so made, that Arabic gold coins, then in free circulation in Syria, would be included, and high ecclesiastics, such as the legates, would easily become possessed of the same, and might preserve them as curiosities; or it may have been that, seeing that the Arab dinar was of a known weight and quality of gold, one of those coins was especially brought to England to fix thereby the standard and quality of ‘the gold pieces’ to be paid as tribute by Offa.
If such was the case, and Offa so received a sample coin, the Mercian king, according to the almost slavish superstitions of that period, would naturally desire to scrupulously perform his vow to the very letter, and to accomplish this object he would have the sample coin faithfully imitated and struck in his own mint, and stamped in addition with his own name and title, ‘OFFA REX,’ in order that no question could thereafter arise as to the exact fulfilment of the vow in regard to the species of coin promised, or as to the identity of the sender of the contribution. That Offa did keep his promise is certain, for in the papal letter sent in 798 by Pope Leo III to Offa’s successor, King Coenwulf, requesting that monarch to continue the donation, it is distinctly so stated.  It may, therefore, be reasonably presumed that Offa’s coin was struck about 787, and was one of the 365 gold pieces sent to Rome in pursuance of his vow.
It is significant to know that this particular coin – so far as we know, the only one now extant – was purchased by a certain Duke de Blacas, an enthusiastic numismatist, in Rome, about a century ago.
No similar coin has been found in England. All this goes to show that the coin was not struck to be put into circulation in England, but was coined for a special purpose, such probably being the payment by Offa of the promised tribute to Rome.
If such be the case, then what bitter irony, unconsciously, accompanied the gift! One of the claims of the Head of the Catholic Christian is that he is the ‘Apostle of God,’ and the ‘Vice-regent of Christ upon the earth.’ Yet, here to his teeth his faithful servitor, Offa, sends as a tribute 365 golden coins, on each of which is plainly stated: There is but One Allah, the Only God, the True, and Muhammad is His Prophet!
Well might Cowper write the lines:
‘God moves in a mysterious way His wonders to perform’!
 This was the period when England was divided into what was termed the Heptarchy, or seven kingdoms. These were: Mercia, Kent, East Anglia, Northumbria, Wessex, Sussex, and Essex. So far as is known, only the first five kingdoms named above struck coins.
 The origin of the Runic writing has been a matter of prolonged controversy. The runes were formerly supposed to have originated out of the Phoenician or the Latin letters, but it is now generally agreed that they must have been derived, about the sixth century B.C., from an early form of the Greek alphabet which was employed by the Milesian traders and colonists of Olbia and other towns on the northern shores of the Black Sea. The Runic alphabet (the oldest of which contained 24 runes, divided into 3 families, each of 8 runes) is called the Futhorc, from the first six letters thereof, f, u, th, o, r, c. The old Norse word run originally meant ‘secret’ or magical. The oldest extant Runic records probably date from the first century of the Christian era, the latest from the fifteenth to the sixteenth century; the greater number are older than the eleventh century, when, after the conversion of the Scandinavians to Christianity, the Futhorc was superseded by the Latin alphabet.
 The letters £ s. d., which are used as abbreviations for pounds, shillings, and pence, owe their origin to certain Latin words used to denote coins. Thus £ signifies libra, a pound sterling; s signifies solidus. The Romans divided their coinage thus: one libra equaled 20 solidi, each solidus being equal to 12 denarii (the denarius thus being, as the modern English penny is to-day, the 240th part of the libra or pound); d signifies denarius, a penny, a word derived from the Latin deni, ten each, from decem, ten. The denarius was the principal silver coin of ancient Rome. The earliest money of Rome was of bronze, and the standard was the as. In 269 B.C. the as was fixed by law at a low valuation, and a silver coin was introduced, with the denarius = to 10 asses, and the quinarius = to 5 asses. The penny (Anglo-Saxon, penig; German, Pfenning, Pfennig) probably derives its name from the Middle Latin word panna, itself derived from the Latina, patina, a shallow bowl. After the sceattae the penny is the most ancient of the English coins, and was the only one current among the Anglo-Saxons. It is first mentioned in the laws of Ina, King of the West Saxons, about the close of the seventh century of the Christian era. It was at that time a silver coin, and weighed about 22.5 troy grains. Halfpence and farthings were not coined in England until the reign of Edward I., but the practice previously prevailed of so deeply indenting the penny with a cross mark that the coin could be easily broken into two or four parts as was required. In 1672 an authorized copper coinage was established in England and halfpence and farthings were struck in copper. The penny was not introduced until 1797, and at the same period the coinage of two-penny pieces was begun; but these latter being found unsuitable were withdrawn. The penny of the present bronze coinage is of only half the value of the old copper coin.
 In Anglo-Saxon times mints for the coinage of money existed at London, Canterbury, and Malmesbury, and coins are extant bearing the names Dorovernis (Canterbury), Londuni (London), and Mealldenus (Malmesbury), showing the place where they were struck.