A group of closely related packs of early woodcut playing cards, among these two tarot packs, survive in the form of several larger and smaller fragments of uncut sheets of cards, as well as a small number of individual cards. Some of the fragments surviving together can reasonably be assumed to be parts of the same sheet, while others are from different imprints from the same woodblock. By combining several fragments, the more or less complete design of many of the woodblocks can be pieced together. From the selection of cards on each block as well as stylistic considerations, it is possible to ascertain which blocks that belong together, forming a complete pack of cards.
The largest collection is located in Szépművészeti Múzeum (Museum of Fine Arts), Budapest. What seems to be duplicates separated from this collection are held by the Metropolitan Museum of Art, New York and the Beinecke Rare Book & Manuscript Library at Yale University, New Haven, Connecticut. In the following, I reference or depict all the fragments I have found documentation on in these and other collections, but I have been informed that there exist further unpublished fragments as well.
The cards are clearly of north Italian origin and from the
renaissance period, but the exact place and time of origin is subject to
debate — Michael Dummett assigns them to Ferrara and dates them to the
second half of the fifteenth century. Their most characteristic feature
is the design of the pip cards in the suit of spade or
swords in all packs where any of these are preserved, which can be
described as intermediate between the
All blocks are of the same size – slightly smaller than a modern A3 sheet. Each consists four rows of cards, with six cards per row in the majority of the blocks. Three blocks from the main collection are notably different in having only five slighly wider cards in each row. Two of these are the only ones to contain tarot trumps; together they contain all trumps and court cards of a standard tarot pack. Given the stylistic similarity and the very peculiar distribution of the trumps between them, it is obvious that they are part of the same pack.
The third of these blocks contains all cards from ace to ten in the suits of bastoni (batons) and spade (swords). Again, stylistical and pragmatical reasons indicate that it is part of the same pack as the previous two. A lost block with the same structure, only with the suits of coppe (cups) and denari (coins), would complete the set of four blocks of 20 cards each for a standard 78-card tarot pack (two positions are left blank in the first block).
Four badly preserved fragments in the private collection of Silvio Berardi in Bologna are depicted in A History of Games Played with the Tarot Pack: The Game of Triumphs by Michael Dummett and John McLeod. These form the bottom and top parts respectively of two sheets exacly like mirror images of the first and third blocks described above. Such reversals are commonplace, as new woodblocks were often made using prints from an older one as a template, but here even the roman numerals on the trumps and the legends on the suit of bastoni are reversed.
In the diagrams, the background colour denote the proportion of the area of each card is preserved when the combination of all fragments are considered:
White – (almost) all
Yellow – three quarters
Peach – half
Pink – a quarter or less
Red – entirely missing
Grey – unknown staus
The titles of the extant blocks link to images of the preserved
fragments. Expansions of the abbreviated card identifiers are
The remaining blocks, all with 24 cards, fall in three subtypes. Type
1 have the twelve court cards – fante (knave), cavallo (knight) and re (king) in each
suit – in the lower half, while the upper half and the type 2 blocks
have the cards from ace to nine in each suit. While this covers the
normal structure for packs of the
Spanish and Portuguese
types of cards, Italian style cards of the period also contains
tens. This discrepancy is explained by the single instance of a type 3
block, consisting of six copies of each of the four tens.
One pair of type 1 and 2 two blocks obviously belong together because
of stylistic similarities and a distinctive division of the non-court
cards between the two blocks. The single surviving type 3 block matches
these (and only these) stylistically, resulting in the only completely
preserved pack in the collection. I have chosen the term
denote this pack, as it differs from the other packs in being somewhat
simpler both in design and execution of the pip cards.
A pair of substantial fragments are located in Museo Civico in Bassano del Grappa. Both contain all or part of every card in a type 1 block, containing the exact same selection of cards as block C1 described above, although in a different ordering. In the top half the designs are largely the same, but there are minor differences in the design of the asso di bastoni and in some of the legends of the bastoni pip cards. In the lower half, however, only the cavallo di spade is of the same design, although all are in a very similar style.
As mentioned above, the
Spanish and Portuguese types of
cards have the same structure as Italian ones, except that they
lack tens. As it is precisely the tens that were relegated to a separate
printing block in the
crude pack, one might speculate that some
packs were made for export, using only the type 1 and 2 blocks. Still,
as the designs are recognisably
Italian (though with a
significant variation in the suit of spade), I find it
probable that a type 3 block once existed also for the packs of which we
now have evidence only for type 1 and sometimes type 2 blocks. The lack
of evidence for such sheets, if they ever existed, is primarily due to
there being printed only a sixth as many of them as the other two types;
a proportion which could be even lower if part of the production was
sold to a market where 48-card packs were preferred. On top of this
comes the circumstance that what survives tends to be sheets that were
never cut up into cards, but was rejected and reused as scrap paper. As
type 1 and 2 sheets became worn at a rate at least six times greater, it
must be assumed that these suffered from misprints leading to rejection
more often as a result.
One group of six fragments is located in Museo Fournier de Naipes (the Fournier Playing Card Museum), Vitoria-Gasteiz. Two fragments overlap, forming the top half of a type 1 block, another consists of an entire lower half of such a block, with a small tab extending into the upper half, and this matches the corresponding part of the upper half. The three remaining fragments do not overlap, but if it is assumed that they belong together, they form an exact type 2 complement to the type 1 sheet preseved alongside it. As they are also stylistically very compatible, there can be no doubt that they form parts of the same pack.
The pip cards from two to nine in the suit of spade are identical to those in the tarot pack, apart from having the right and left edges trimmed somewhat in order to fit on the narrower cards, and those of bastoni are very similar. While the knaves and knights have unique designs in the general style of the entire group, the kings are very similar to those in C1, although different enough that they cannot be copies of the same original design.
In the main collection there is a type 1 block where all but one of
the cards in the upper half are mirror-image copies of the corresponding
Fournier designs. Such reversals are common in historical card
production, as the easiest way to copy a pack was to trace an existing
set of cards directly onto the face of a woodblock, which of course
gives a reversed image when used for printing. What is remarkable about
this instance (like the mirror-image version of the Tarot sheets) is
that the text denoting the value of the pip cards in the suit of bastoni is also reversed. For this reason I have chosen to
call this the
reversed pack. Its court cards do not copy the same
Fournier sheet, however, but two of the knight cards are mirror-image
copies of those in the
crude pack, and a third is a non-reversed
The lost type 2 and 3 blocks would presumably be mirror images of the corresponding blocks of the Fournier pack. However, as the asso di denari (ace of coins) is different, it is possible that also the other cards of this suit also had another design.
Four fragments out of a group of five previously in the private
collection of Giulio Bernardi of Trieste contains between them almost
complete imprints from two type 1 blocks that are near perfect mirror
images of each other. The only publication of these to my knowledge is
an entry by Alberto Milano in a 1984 catalogue, where only the fragments
making up the lower halves of the sheets containing the court cards are
depicted; the upper halves being only briefly described. With the
exception of the kings being identical to those in C1, the lower halves
are very close copies of R1, both in layout and individual card designs.
From the description, it is clear that the same must be the case for the
upper halves. In the table, I have therefore suggested that the layout
of the sheet with non-reversed text is laid out like a mirror image of
R1, even though the description lists 4 to 9 of batons in ascending
order. The legend on the two of batons is given as
obviously a misreading of
duob[us] as in the
Fragment 5 contains a cardback design which according to another description matches the one I hypothesised for the main group.
The relationship between the final block and all the others presents something of a conundrum.
On one hand, these cards are stylistically less closely related to
the other packs, and shares no identical designs with any of them. It
has more detailed pip cards in the suit of coppe than
the Fournier pack, which in this regard is substantially better than the
crude pack (the only other packs to preserve these cards). The
court cards are more refined, using hatched shading (this is also used
in the Fournier pack) and more detailed foregrounds in the court cards.
I have therefore chosen to call this the
On the other hand, it has the same block and card sizes, identical
borders around the cards and around the outer edge. The selection of
cards and their general placement within the block is identical to block
1 of the Fournier and
reversed packs, except having the pip cards
from two to nine of coppe instead of bastoni. It is not only preserved in the same collection
as the majority of the other sheets, but the chararteristic shapes of
the fragments, including small rectangular holes along one of the short
edges, indicates that it went through the same kind of reuse after being
scrapped. Thus these fragments are not merely placed among the others at
a later date by a collector.
Absent any other explanation of the latter points, my tentative conclusion is that despite the stylistic differences, this pack was most likely made in the same workshop and at the same time as the others.
On this background, it is of great interest to observe that the
elegant pack’s asso di spade (ace of swords)
fits much better with the pip cards of this suit in the tarot and the
Fournier packs than the ones actually present in those packs.