Though a great deal has been written about minting processes old and new, some of the biggest remaining gaps in our knowledge of them are connected with the various ways in which the special mintages called "master coins" or "proofs" have been made over the centuries, why changes in finish have occurred, or indeed just when such coins started to be made and why. I hope to be able to throw a little light on all these questions, though the full answers are not yet available.
Any such study, even though emphasizing principally American Colonial, U.S., and Canadian coins, cannot be limited to them, because our mint personnel obtained their knowledge of minting processes (including those connected with proofing) from other sources, primarily the Royal British Mint on Tower Hill, London, and later on Boulton & Watt's Soho Mint near Birmingham, England -and more recently from other European sources. It was for just such reasons, for instance, that in the 1830's the Philadelphia Mint financed a transatlantic trip for Franklin Peale during which he visited all the principal European mints. This is why I shall go to what might seem unusual length tracing changing methods of making proof coins in Britain. But there are even more obvious reasons: some Colonial and early u.s. proofs - now hardly ever obtainable for study -were made by methods virtually identical to those used on the far less rare or costly British proofs, and study of the latter makes recognition and understanding of the former far more easy. And most Colonial proofs were made at the Tower Hill mint, (e.g. the 1773 Virginia "Penny", 1774 Virginia Shilling), or under supervision of British mint officials elsewhere in London or Bristol (including all the Rosa Americana and Wood's Coinage proofs), or at the Boulton & Watt mint (the 1796 Kentucky Myddelton patterns). Furthermore, the remark holds for both pre-decimal and decimal series Canadian proofs; all of the decimal ones until the beginning of the 20th century were in fact made in London.
Since the term "proof" has shifted its meaning, even as the coins so designated have changed in function over the decades and centuries, this study has to pay some attention to the shifting meanings and purposes. The very term "proof" seems to have been of British origin, analogous at the start to its use in graphic arts (i.e. preliminary trial, as of an engraving plate), and may at inception have meant something very much akin to an artist's proof piece -something struck from master dies ordinarily used only for making hubs, and too precious to be risked directly on production coinages where breakage would necessitate an immediate halt and perhaps weeks of delay while another such die was being perfected.
Later on, and most familiarly, the name "proof" began to be applied to various kinds of special "polished-up portraits of the coinages" (Sheldon's term), presentation pieces remotely like business strikes in appearance or finish, though of normal designs, and made by different processes primarily intended for medals and calculated to show off what splended results moneyers could achieve if they paid particular attention to each individual piece, principally in improved relief detail. This usage seems to have been introduced to American numismatics in 1858 by James Ross Snowden; the Oxford English Dictionary does not record the term in this sense (*F14) before 1901!
More recently, of course, the name "proof" has become merely a technical term for a kind of mass-production coins with unusually shiny surfaces, things of curiosity value, little if at all more attractive than the regular coinage, intended less for display or presentation purposes than for sale to investors - often enough (through the 1950's) being bought, sold and traded in quantity in mint sealed boxes without even having been looked at. Together with this deterioration in meaning and function has come similar deterioration in appearance of the coins, enormous quantity and stereotypy, and great overvaluation as well.
Nothing comparable to proofs seems to have been known in antiquity, though Greek technology permitted the striking of coins in far higher relief than is practicable today, and Greek celators created individual dies of artistic merit ranging from fairly good to extraordinary. Why nobody thought of polishing dies or blanks is unknown; probably this refinement was thought unnecessary, the high relief affording ample contrast between devices and fields.
In later centuries, until the Renaissance, despite excellence of design and execution of some dies, such characteristic features of later proofs as high relief and unusually sharp striking remained impossible while moneyers used the ancient hammer method of striking. Even several blows from the largest sledge-hammers hardly sufficed to bring up designs in more than Slight relief without very marked danger of shattering the dies. Multiple strikings (as nearly always on multiple thalers and other very large coins) usually tended to impair the general appearance of the coins because successive blows imparted confusing or mutually obliterative extra lines to each letter or design element. This is easily enough understood: after all, planchets - not being confined by collars - spread out from each successive blow, often irregularly to the point of no longer being round.
It follows, then, that we need not seek a beginning to the proofing process earlier than the 15th century, when Italian and French medallists began experimenting with "Coyning Engines" (as their British successors called the apparatus) ancestral to the 18th Century screw-press and on the same principle. The advantages these rotating pile-drivers had over the hammer method were largely those of producing a more vivid impression from dies in higher relief; the Italian medallists who invented the screw press did so in frank efforts to make medals in the manner of Roman coins. An immediate by-product of this technological innovation was that dies could be hubbed, saving a great deal of time and handwork. Placing a die blank in the press opposite a carved and hardened relief model or hub meant that one could immediately impart devices to a working die which might otherwise have taken weeks of the most eye-taxing labor; further, one could multiply similar dies and accommodate even the largest coinages, insuring meanwhile against delays due to die breakage.
I have not been able to learn who introduced the practice of polishing dies and blanks and making multiple impressions to bring out details in unusual clarity. Some sort of special proofing treatment dates back to the early 1660's; earlier coins alleged to show it are very equivocal, even British experts sometimes being unable to ascertain for certain whether the pieces in question were made by these medallic processes. Had they been in France in the 1500's, most probably the French fugitive Eloye Mestrell would have produced some pieces by them during his own ill-fated experiments in the Tower Mint - as gifts to the British monarch, showing that whether or not his coinages were slower than those of the regular moneyers they nevertheless showed designs of unusual merit, clarity and vividness. No such pieces seem to be known.
Seaby's monograph, English Silver Coins, lists only two pieces prior to 1662 with any claim to proof status, and about both there is doubt. One of these is ESC 427, a 1651 Commonwealth half-crown, with footnote (p. 47) "May be just an exceptionally well struck ordinary coin." The other is a 1658 Cromwell half-crown struck in gold, from dies by the illustrious Thomas Simon, ESC 447a. Now Seaby generally refers to off-metal impressions (favor coins, pieces de caprice, and the like) as "proofs," without any remark on their method of manufacture; this may be mere linguistic usage, following OED, but if so it is confusing indeed. Never having seen the gold Cromwell piece in question, I cannot say if it was made by processes comparable to those used on later British proofs.