Phaistos Disc

The disk is made of fine-grained clay. Some authors have stated that the clay does not appear to be of local origin, perhaps not even from Crete.[6] It was intentionally and properly fired, unlike tablets and seals that were baked only accidentally, by building fires.[6]

The disk is approximately cylindrical, about 16 cm in diameter and almost 2 cm thick, with rounded edges. More precisely, the outline is slightly egg-shaped, with the diameter varying from 15.8 to 16.5 cm and the thickness from 1.6 to 2.1 cm. The disk is slightly concave on side A and convex on side B.[6]

The most remarkable feature of the Phaistos disk is that the embossed signs that comprise its inscription all result from pressing separate stamps – one for each symbol – into the soft clay before firing. Thus the disk can be seen as an early example of movable-type printing.[7][8] Typesetter and linguist Herbert Brekle writes:[9]

If the disc is, as assumed, a textual representation, we are really dealing with a "printed" text, which fulfills all definitional criteria of the typographic principle. The spiral sequencing of the graphematical units, the fact that they are impressed in a clay disc (blind printing!) and not imprinted are merely possible technological variants of textual representation. The decisive factor is that the material "types" are proven to be repeatedly instantiated on the clay disc.

A medieval example of a similar blind printing technique[10] is the Prüfening dedicatory inscription.[9][11]

Popular-science author Jared Diamond describes the disc as an example of a technological innovation that did not become widespread because it was made at the wrong time in history. Diamond contrasts the process with Gutenberg's printing press.[12]

Besides the stamped symbols, there are a few markings made by scoring the moist clay with a sharp stylus. On each side there is a continuous spiral line that separates successive turns of the text. The strip between successive spires of this line is divided into sections by short radial lines, so that each section contains a few whole signs. The presumed start of the text, adjacent to the edge, is also marked by such a radial stroke, with the addition of five dots punched along it with the stylus. Finally, under some of the stamped signs, there are short oblique strokes.

There are 45 distinct signs on the disk, occurring a total of 242 times — 123 on side A and 119 on side B. In addition to these, a small diagonal line was incised with a stylus (not stamped) underneath some signs, a total of 18 times. The 45 symbols were numbered by Arthur Evans from 01 to 45,[1] and this numbering has been adopted by most researchers.

The signs were added to the Unicode universal computer character (UCS) set in 2008, after a 2006 proposal by Michael Everson and John H. Jenkins.[13] In the following table, the No. column is the Evans number of each sign; the Glyph column is a modern drawing of the symbol; and the Font column uses the UCS font available in the browser. The assigned Unicode names are PHAISTOS DISC SIGN followed by the names shown under Name in the table below, taken from a 1995 book by Louis Godart.[14]

One sign occurrence on side A is too damaged to identify. According to Godart, it may be sign 03 (TATTOOED HEAD) or 20 (DOLIUM); or less probably 08 (GAUNTLET) or 44 (SMALL AXE).[14]^: p.101 Theoretically, it could also be a 46th distinct sign.

The sign images below are reversed left-to-right relative to their appearance on the disk, reflecting their presentation in most Western books and articles.[15]

Also, some signs occur in the disk in two or more orientations, rotated by 90 or 180 degrees. It is generally assumed that the rotation has no semantic or linguistic value, so the rotated copies are still the same symbol. Therefore, the "normal" orientation of those signs is not known, and might have been left to the scribe's discretion.[13][15]

Sign				Frequency
No.	Glyph	Font	Name	A	B	A+B
01		𐇐	PEDESTRIAN	6	5	11
02		𐇑	PLUMED HEAD	14	5	19
03		𐇒	TATTOOED HEAD	2	0	2
04		𐇓	CAPTIVE	1	0	1
05		𐇔	CHILD	0	1	1
06		𐇕	WOMAN	2	2	4
07		𐇖	HELMET	3	15	18
08		𐇗	GAUNTLET	1	4	5
09		𐇘	TIARA	0	2	2
10		𐇙	ARROW	4	0	4
11		𐇚	BOW	1	0	1
12		𐇛	SHIELD	15	2	17
13		𐇜	CLUB	3	3	6
14		𐇝	MANACLES	1	1	2
15		𐇞	MATTOCK	0	1	1
16		𐇟	SAW	0	2	2
17		𐇠	LID	1	0	1
18		𐇡	BOOMERANG	6	6	12
19		𐇢	CARPENTRY PLANE	3	0	3
20		𐇣	DOLIUM	0	2	2
21		𐇤	COMB	2	0	2
22		𐇥	SLING	0	5	5
23		𐇦	COLUMN	5	6	11
24		𐇧	BEEHIVE	1	5	6
25		𐇨	SHIP	2	5	7
26		𐇩	HORN	5	1	6
27		𐇪	HIDE	10	5	15
28		𐇫	BULLS LEG	2	0	2
29		𐇬	CAT	3	8	11
30		𐇭	RAM	0	1	1
31		𐇮	EAGLE	5	0	5
32		𐇯	DOVE	2	1	3
33		𐇰	TUNNY	2	4	6
34		𐇱	BEE	1	2	3
35		𐇲	PLANE TREE	5	6	11
36		𐇳	VINE	0	4	4
37		𐇴	PAPYRUS	2	2	4
38		𐇵	ROSETTE	3	1	4
39		𐇶	LILY	1	3	4
40		𐇷	OX BACK	3	3	6
41		𐇸	FLUTE	2	0	2
42		𐇹	GRATER	0	1	1
43		𐇺	STRAINER	0	1	1
44		𐇻	SMALL AXE	1	0	1
45		𐇼	WAVY BAND	2	4	6

Many of signs are depictions of concrete objects with a recognizable general nature (such as humans, birds, plants, a boat), or parts thereof (heads, hides, flowers). However, in most cases the precise nature of objects depicted is still unknown (as of 2023). The sign names assigned by scholars, in particular by Godart[14] and the Unicode consortium,[13] were rather arbitrary, often based on the slightest shape similarity.

Symbol 21 (Godart's "COMB") was once conjectured to be a palace floorplan.[16] However, this hypothesis was cast in doubt by the discovery of a vase with a nearly identical symbol incised on the bottom, believed to be a potter's mark.[17]

Ritual sea snail (triton) conch, decorated with red paint. Phaistos, 3600–3000 BC.

Symbol 20 ("DOLIUM") was assumed to be the conch of a large sea snail, such as Tonna dolium or some Eudolium species. One such conch was found at Phaistos and is believed to have been used as a musical instrument for ritual uses.

The distribution of symbols is highly non-random, and quite different between the two sides. Evans's symbol 02 (PLUMED HEAD) is most frequent one, occurring 19 times — 14 of them on side A. The next most frequent signs are 07 (HELMET), with 18 occurrences, mostly on side B; 12 (SHIELD), with 17, mostly on side A; and 27 (HIDE), with 15, of which 10 on side A.

Still, 26 of the 45 symbols occur on both sides, at least once on each. The most common signs that occur on only one side are 31 (EAGLE), on side A, and 22 (SLING), on side B; both with 5 occurrences each.

The following table shows how many distinct signs (Sign count) have the same number of occurrences (Frequency). The third number in each column is the product of the two numbers above, that is, the total number of occurrences (Token count) of those signs:

The nine hapaxes (symbols occurring just once) are 04 (A5), 05 (B3), 11 (A13), 15 (B8), 17 (A24), 30 (B27), 42 (B9), 43 (B4), 44 (A7). Of the eight twice-occurring symbols, four (03, 21, 28, 41) occur on side A only, three (09, 16, 20) on side B only, and only one (14) occurs on both sides.

The distribution of symbol pairs too is highly non-uniform. For example, of the 17 occurrences of sign 12 (SHIELD), thirteen follow immediately sign 02 (PLUMED HEAD).

Evans, at one point, published an assertion that the disc had been written, and should be read, from the center out; because it would have been easiest to place the inscription first and then size the disc to fit the text. There is general agreement that he was wrong, and Evans himself later changed his mind: the inscription was made, and should be read, in the clockwise sense, from the outside in toward the centre, as with the similar spiral inscription on the Lead Plaque of Magliano.[18]^: 649 Yves Duhoux stated that any outward reading may be discarded.

Among the arguments that support this conclusion is the fact that the centres of the spirals are not in the centre of the disc. Also, some of the symbols near the centre are crowded, as though the maker was cramped for space. One pair of symbols is even stacked in the "vertical" (radial) direction rather than side-by-side. Furthermore, whenever symbols overlap the inner symbol overlies the outer symbol. (However, this analysis is complicated by several apparent corrections made by the scribe, where the stamping of the corrected symbols may well have been different from the reading order.) Jean Faucounau has proposed a reconstruction of the scribe's movements, which would also require an inward direction.

The outside-to-centre reading direction implies right-to-left reading of the text, which is into the faces of the people and animals — as is the case in Egyptian and Anatolian.

In spite of this widespread consensus, some authors still dispute this conclusion and attempt to decipher it under the opposite reading order (left-to-right, counterclockwise, from the center to the periphery).[19]

Phaistos disk scholars publishing in languages with left-to-right writing systems have generally found it convenient to mirror the text so that it reads from left to right. In this case, for consistency, the individual symbols are mirrored as well. (This practice is commonly seen also in many publications about Egyptian and Anatolian hieroglyphic texts.) The Unicode character set was designed to be used in this way, thus Unicode fonts typically depict the characters flipped relative to their orientation on the disk.

Evans considered side A as the "front" side, but technical arguments have since been forwarded favouring side B as the "front" side.

The following is a single image of the text "unrolled". While the order of the characters is left-to-right reversed, the signs themselves are in the original orientation.

Unrolled and left-to-right reversed image of the text.

The signs are laid out on each side as a single spiral text, which is split by the inscribed radial strokes into "words". Both ends of the text on each side are also assumed to be "word" boundaries. There are 61 such "words" on the Disc, with two to seven sign occurrences each: 31 on side A and 30 on side B. These "words" are conventionally numbered A1 to A31 and B1 to B30, reading from right to left (clockwise, edge-to-center).[14] The signs are laid out right next to the disk's edge at first (13 words on A, 12 words on B) before spiralling toward the center (18 more words on each side).

There may be one additional radial stroke near the center of side A, over-stamped by the sign 03 (TATTOOED HEAD), between sign 10 (ARROW) and the central sign 38 (ROSETTE). However, most scholars ignore that possible stroke and count the last three symbols as a single "word" 10-03-38 (which happens to occur also at about the same position on the next-to-last turn).[14]

On both sides, there is a radial line also right before the start (outermost end) of the text, with five dots punched along it using a sharp round stylus.

Fields numbering by Louis Godart

The short oblique strokes that were drawn with a stylus (not stamped) below some signs are always attached to the last sign of a "word" (assuming outside-in reading direction). Their meaning is a matter of discussion. One hypothesis, supported by Evans,[1] Duhoux, Ohlenroth and others, is that they further subdivide the text into "paragraphs". However, alternative meanings have been offered by other scholars.

Most transcriptions have 10 of those oblique strokes on side A and 8 on side B. That mark occurs at the end of the last word on side B, but not on side A; however, the end of the text on side A is generally assumed to end a "paragraph" as well. There are thus 11 and 8 "paragraphs", respectively, on sides A and B. Those 19 "paragraphs" have between one and four "words", with two exceptions: one "paragraph" on side A with nine words (A4 to A12), and one on side B with 12 words (B7 to B18).

However, oblique strokes at the end of words A7 and B8 may have been lost due to damage along the edge of disk; and the discontinuity in the placement of symbols between B12 and B13, following the end of the first turn of the text, might have served as a "paragraph" break too.

Also, close inspection of modern high-resolution photographs of the disk show that the "stroke" commonly listed under sign 19 at the end of word A24 is not really a stroke sign, but merely part of a crack that extends from the center of the disk to its edge, cutting through several signs. Excluding that occurrence, there would be only 9 oblique strokes and 10 paragraphs on side A, and words A23 to A27 would be a single five-word paragraph.

The following transcriptions of the text all assume a right-to-left (clockwise, edge-to-center) reading direction on the disk, starting at the vertical (radial) line of five dots. In these transcription, however, the order of the characters has been flipped, so that they should be read left-to-right and top-to-bottom. The oblique stroke is assumed to indicate the last word of a "paragraph". A horizontal line has been added after each "paragraph" for clarity.

For consistency with most published sources, these transcriptions assume that there is an oblique stroke at the end of word A24, even though high-resolution images show it to be just a crack.

The following is a rendering of the Phaistos Disc inscription in Unicode characters from the Phaistos code block (code points +101D0 to +101FC). The radial strokes are denoted by the ASCII character "|" (+7C), and the oblique subscripted stroke by the comma-like PHAISTOS DISK COMBINING OBLIQUE STROKE (+101FD) after the affected symbol. The radial stroke with five dots, that indicates the presumed start of text, is denoted by the ISO Latin 1 character "¦" (+A6). The boxed question mark "⍰" (+2370) denotes the illegible sign in word A8. The appearance of the signs will depend on the font used by the browser, but normally they should all be left-to-right flipped relative to their appearance on the disk.

Side A:

¦ 𐇑𐇛𐇜𐇐𐇡𐇽

---

| 𐇧𐇷𐇛 | 𐇬𐇼𐇖𐇽

---

| 𐇬𐇬𐇱 | 𐇑𐇛𐇓𐇷𐇰 | 𐇪𐇼𐇖𐇛 | 𐇪𐇻𐇗 | 𐇑𐇛𐇕𐇡⍰ | 𐇮𐇩𐇲 | 𐇑𐇛𐇸𐇢𐇲 | 𐇐𐇸𐇷𐇖 | 𐇑𐇛𐇯𐇦𐇵𐇽

---

| 𐇶𐇚 | 𐇑𐇪𐇨𐇙𐇦𐇡 | 𐇫𐇐𐇽

---

| 𐇑𐇛𐇮𐇩𐇽

---

| 𐇑𐇛𐇪𐇪𐇲𐇴𐇤 | 𐇰𐇦 | 𐇑𐇛𐇮𐇩𐇽

---

| 𐇑𐇪𐇨𐇙𐇦𐇡 | 𐇫𐇐𐇽

---

| 𐇑𐇛𐇮𐇩𐇽

---

| 𐇑𐇛𐇪𐇝𐇯𐇡𐇪 | 𐇕𐇡𐇠𐇢𐇽

---

| 𐇮𐇩𐇛 | 𐇑𐇛𐇜𐇐 | 𐇦𐇢𐇲𐇽

---

| 𐇙𐇒𐇵 | 𐇑𐇛𐇪𐇪𐇲𐇴𐇤 | 𐇜𐇐 | 𐇙𐇒𐇵

Side B

¦ 𐇑𐇛𐇥𐇷𐇖 | 𐇪𐇼𐇖𐇲 | 𐇑𐇴𐇦𐇔𐇽

---

| 𐇥𐇨𐇪 | 𐇰𐇧𐇣𐇛 | 𐇟𐇦𐇡𐇺𐇽

---

| 𐇜𐇐𐇶𐇰 | 𐇞𐇖𐇜𐇐𐇡 | 𐇥𐇴𐇹𐇨 | 𐇖𐇧𐇷𐇲 | 𐇑𐇩𐇳𐇷 | 𐇪𐇨𐇵𐇐 | 𐇬𐇧𐇧𐇣𐇲 | 𐇟𐇝𐇡 | 𐇬𐇰𐇐 | 𐇕𐇲𐇯𐇶𐇰 | 𐇑𐇘𐇪𐇐 | 𐇬𐇳𐇖𐇗𐇽

---

| 𐇬𐇗𐇜 | 𐇬𐇼𐇖𐇽

---

| 𐇥𐇬𐇳𐇖𐇗𐇽

---

| 𐇪𐇱𐇦𐇨 | 𐇖𐇡𐇲 | 𐇖𐇼𐇖𐇽

---

| 𐇖𐇦𐇡𐇧 | 𐇥𐇬𐇳𐇖𐇗𐇽

---

| 𐇘𐇭𐇶𐇡𐇖 | 𐇑𐇕𐇲𐇦𐇖 | 𐇬𐇱𐇦𐇨 | 𐇼𐇖𐇽

---

The following transcription uses modern drawings of the signs, which are left-to-right reversed with respect to their appearance on the disk. The labels A1-A31 and B1-B30 are the traditional word numbers.[14]

Side A:

^A1

---

^{A2 A3}

---

^{A4 A5 A6 A7 A8 A9 A10 A11 A12}

---

^{A13 A14 A15}

---

^A16

---

^{A17 A18 A19}

---

^{A20 A21}

---

^A22

---

^{A23 A24}

---

^{A25 A26 A27}

---

^{A28 A29 A30 A31}

Side B:

^{B1 B2 B3}

---

^{B4 B5 B6}

---

^{B7 B8 B9 B10 B11 B12 B13 B14 B15 B16 B17 B18}

---

^{B19 B20}

---

^B21

---

^{B22 B23 B24}

---

^{B25 B26}

---

^{B27 B28 B29 B30}

---

The following transcription uses the Evans numbers for the signs.[5] The vertical bar characters "¦" and "|" represent the start-of-text and word-separating radial lines, respectively. The slash "/" denotes the oblique stroke under the preceding sign. The caret "^" indicates the transition from the first turn of the text (along the disk's edge) to the inner turns, and "??" is the unreadable sign.

Side A:

¦ 02 12 13 01 18/

---

| 24 40 12 | 29 45 07/

---

| 29 29 34 | 02 12 04 40 33 | 27 45 07 12 | 27 44 08 | 02 12 06 18 ?? | 31 26 35 | 02 12 41 19 35 | 01 41 40 07 | 02 12 32 23 38/

---

| 39 11 | ^ 02 27 25 10 23 18 | 28 01/

---

| 02 12 31 26/

---

| 02 12 27 27 35 37 21 | 33 23 | 02 12 31 26/

---

| 02 27 25 10 23 18 | 28 01/

---

| 02 12 31 26/

---

| 02 12 27 14 32 18 27 | 06 18 17 19/

---

| 31 26 12 | 02 12 13 01 | 23 19 35/

---

| 10 03 38 | 02 12 27 27 35 37 21 | 13 01 | 10 03 38

Side B

¦ 02 12 22 40 07 | 27 45 07 35 | 02 37 23 05/

---

| 22 25 27 | 33 24 20 12 | 16 23 18 43/

---

| 13 01 39 33 | 15 07 13 01 18 | 22 37 42 25 | 07 24 40 35 | 02 26 36 40 | 27 25 38 01 | 29 ^ 24 24 20 35 | 16 14 18 | 29 33 01 | 06 35 32 39 33 | 02 09 27 01 | 29 36 07 08/

---

| 29 08 13 | 29 45 07/

---

| 22 29 36 07 08/

---

| 27 34 23 25 | 07 18 35 | 07 45 07/

---

| 07 23 18 24 | 22 29 36 07 08/

---

| 09 30 39 18 07 | 02 06 35 23 07 | 29 34 23 25 | 45 07/

---

The following table shows the distinct "words" of the text, their frequencies, and their locations. (The oblique "paragraph separator" strokes are not considered part of the words.)

The up/down arrow buttons on the first column can be used to sort the "words" lexicographically based on the Evans sign numbering — so that all the "words" that start with the same symbols are listed together.

The arrow buttons on the third column can be used to sort the "words" by back-to-front lexicographic order, in which the signs of each word are reversed; so that all the "words" with same ending are listed together. Column 4 still shows each "word" in the presumed reading order, only right-justified to make the common endings more evident.

Normal lex order		Reversed lex order
Codes	Glyphs	Codes (Rev.)	Glyphs	Freq.	Position(s)
01-41-40-07		07-40-41-01		1	A11
02-06-35-23-07		07-23-35-06-02		1	B28
02-09-27-01		01-27-09-02		1	B17
02-12-04-40-33		33-40-04-12-02		1	A5
02-12-06-18-??		??-18-06-12-02		1	A8
02-12-13-01		01-13-12-02		1	A26
02-12-13-01-18		18-01-13-12-02		1	A1
02-12-22-40-07		07-40-22-12-02		1	B1
02-12-27-14-32-18-27		27-18-32-14-27-12-02		1	A23
02-12-27-27-35-37-21		21-37-35-27-27-12-02		2	A17,A29
02-12-31-26		26-31-12-02		3	A16,A19,A22
02-12-32-23-38		38-23-32-12-02		1	A12
02-12-41-19-35		35-19-41-12-02		1	A10
02-26-36-40		40-36-26-02		1	B11
02-27-25-10-23-18		18-23-10-25-27-02		2	A14,A20
02-37-23-05		05-23-37-02		1	B3
06-18-17-19		19-17-18-06		1	A24
06-35-32-39-33		33-39-32-35-06		1	B16
07-18-35		35-18-07		1	B23
07-23-18-24		24-18-23-07		1	B25
07-24-40-35		35-40-24-07		1	B10
07-45-07		07-45-07		1	B24
09-30-39-18-07		07-18-39-30-09		1	B27
10-03-38		38-03-10		2	A28,A31
13-01		01-13		1	A30
13-01-39-33		33-39-01-13		1	B7
15-07-13-01-18		18-01-13-07-15		1	B8
16-14-18		18-14-16		1	B14
16-23-18-43		43-18-23-16		1	B6
22-25-27		27-25-22		1	B4
22-29-36-07-08		08-07-36-29-22		2	B21,B26
22-37-42-25		25-42-37-22		1	B9
23-19-35		35-19-23		1	A27
24-40-12		12-40-24		1	A2
27-25-38-01		01-38-25-27		1	B12
27-34-23-25		25-23-34-27		1	B22
27-44-08		08-44-27		1	A7
27-45-07-12		12-07-45-27		1	A6
27-45-07-35		35-07-45-27		1	B2
28-01		01-28		2	A15,A21
29-08-13		13-08-29		1	B19
29-24-24-20-35		35-20-24-24-29		1	B13
29-29-34		34-29-29		1	A4
29-33-01		01-33-29		1	B15
29-34-23-25		25-23-34-29		1	B29
29-36-07-08		08-07-36-29		1	B18
29-45-07		07-45-29		2	A3,B20
31-26-12		12-26-31		1	A25
31-26-35		35-26-31		1	A9
33-23		23-33		1	A18
33-24-20-12		12-20-24-33		1	B5
39-11		11-39		1	A13
45-07		07-45		1	B30

The distribution of the word-initial symbols is quite non-uniform. In particular, the most common word-initial symbol, 02 PLUMED HEAD, occurs only in that position, and begins 19 of the 61 words. On 13 of its 19 occurrences, it is followed by 12 SHIELD (which however also occurs a few times at the end of words). The next most common word-initial signs are 29 CAT with seven occurrences in that position, 27 HIDE with five, 07 HELMET with four, and 22 SLING with three.

The distribution of word-final signs is also quite non-uniform and quite distinct from that of word-initial signs. The most common word-ending symbols are 07 HELMET, with eight occurrences; 01 PEDESTRIAN, with seven; 35 PLANE TREE with seven; 18 BOOMERANG with five; and then 08 GAUNTLET, 25 SHIP, 26 HORN, and 33 TUNNY, each with three occurrences each.

Six entire words occur twice each:

The three-word sequence "| 02 27 25 10 23 18 | 28 01/ | 02 12 31 26/ |" occurs twice, as words A14-16 and A20-22. Its last word "| 02 12 31 26/ |" recurs for a third time as word A19. Four more words occur twice each, "| 02 12 27 27 35 37 21 |" (A17, A29), "| 10 03 38 |" (A28, A31), "| 22 29 36 07 08/ |" (B21, B26), and "| 29 45 07/ |" (A3, B20).

The disc shows signs of corrections having been made, with some signs erased and over-printed by other signs.

Godart describes these corrections as occurring in the following words: A1 (signs 02-12-13-01), A4 (29-29-34) together with A5 (02-12-04), A8 (12), A10 (02-41-19?-35), A12 (12), A16 (12-31-26?), A17 (second 27?), A29 (second 27?), B1 (12-22), B3 (37?), B4 (22-25 imprinted over the same), B10 (07?-24?-40?), B13 (beside 29?). Question marks indicate uncertainty about that particular sign being the result of a correction.[14]^: p.99-109

The borders of word B28 were also widened to make room for sign 02.[14]^: p.107[4]^: p.34-35

The two signs 27 (HIDE) in word A29 are rotated 180 degrees compared with all other occurrences of this sign: "head down" versus "head up". This rotation might be motivated by lack of space in A29.[4]^: p.24

The rotations of the signs 29 (CAT) and 31 (EAGLE) have no lack of space. Defining the sign 29 in words B19, B20 and B21 as "head to the right", this presents as: head down in B29; head to the left in A3 and B15; head up in B18 and B26; head in between up and left in B13; head in between right and down in A4 (twice). The direction of the head of sign 31 is as follows: to the right in A16, up in A9 and A25, and to the left in A22.

The sign 02 (PLUMED HEAD) in word A29 is 90 degrees rotated to the right compared with all other occurrences of this sign. This might well be due to lack of space; the word is crowded and messy, with the sign 12 (SHIELD) pushed aside.

The two occurrences of sign 28 (BULLS LEG) are not rotated compared with each other. Rather, the way this sign is shown in the literature (including Unicode), with the foot down, is rotated compared with the sign on the disc, with the foot up.

Arie Cate observed that if signs rotations were random with uniform distribution, then the probability that they end up in only two (or three) signs is very small.[20]

There are several locations on side A where two occurrences of the same sign lie near each other in adjacent turns of the spiral, such as sign 02 (PLUMED HEAD) in word A1 and in word A14. Also the two 27 signs (HIDE) signs in word A29 are upside down, with the "heads" pointing to the HIDE sign of word A23, in the adjacent turn. Arie Cate claims that the probability of these alignments being coincidental is rather small.[21]

For the first few decades after its discovery most scholars argued strongly against the local origin of the artifact. Evans[1]^: p.24 wrote that:

...when one comes to compare the figures in detail with those of the Minoan hieroglyphic signary, very great discrepancy is observable... Out of the forty-five separate signs on the Phaistos Disk, no more than ten more or less resemble Cretan hieroglyphic forms... The human figures in their outline and costume are non-Minoan... The representation of the ship also differs from all similar designs that occur either among the hieroglyphic or the linear documents of Crete.

Gustave Glotz claimed that the clay was not from Crete.[22]^: p.381 Ipsen concluded that the disc was certainly from somewhere on the Aegean; however, because of its differences from Linear A or B, he, like Evans, supported a non-Cretan origin for the Disc. He observes, however, that since Linear A was a common Aegean script such an assumption will not resolve the problem of multiplicity.[23]^: p.15

Vase from Knossos with a stamped sign similar to SHIELD.

However, the consensus on this question changed in later decades, as a few other artifacts were found on Crete with significant resemblances to the disk. For instance, a vase found at Knossos (Vase 14 236) bears a stamped sign identical to the disk's 25 SHIELD sign (a circle with seven dots). Also, under the bottom of a bowl found in 1965 at Phaistos (bowl F 4718 from the House South of the Ramp) there is a sign in relief, believed to be a potter's mark, that is basically identical to sign 21 (COMB).[17] A very similar sign is found as an impression on a sealing from a deposit of administrative documents discovered in 1955, beneath Room 25 of the Second Palace of Phaistos (sealing CMS II.5, n. 246).[17][24][25] Female images with pendulous breasts have also been found at Malia and Phaistos.[14]^: p.125 And the Arkalochori Axe bears a short inscription that uses several signs similar to those of the disk.[26]

These and other finds have made Cretan origin more popular.[17] This view was expressed by Michael Trauth in 1990,[27] Duhoux in 2000,[28] Andrew Robinson in 2008.[29]

The uniqueness of the script, of the spiral arrangement, and of the method of writing (individual glyph stamps) have led some scholars to raise the possibility that the Phaistos disc is a 1908 forgery or hoax.[6][30] It was pointed out that the date of manufacture has never been established by thermoluminescence dating.[31] However, the Disc is now generally accepted as authentic by archaeologists.[32] Andrew Robinson concurs that thermoluminescence dating would be highly desirable, but does not endorse the forgery arguments.[29]

The precise excavation records maintained by Luigi Pernier have always been a problem for the hoax hypothesis. That hypothesis was eventually put to rest by the discovery of the other artifacts in Crete with similar glyphs, which a 1909 hoaxer would not have known about. Also, a gold signet ring from Knossos (the Mavro Spilio ring), found in 1926, contains a Linear A inscription laid out in a spiral, similar to the Phaistos Disc.[33]

While the Phaistos disk writing system is, on the whole, very different from other known scripts, several scholars have argued against it being an entirely independent invention. Gunther Ipsen argued that the creator must have been influenced by other scripts, and points out the Hieroglyphic Luwian script from Anatolia as an example of an original script inspired by other writing systems (its symbol values inspired by cuneiform, its shapes by Egyptian hieroglyphs).[23]^: p.11

Several scholars have proposed that the Phaistos signs are older or alternate forms of Linear A glyphs, specifically. Others have pointed to similar resemblances with the Anatolian (Luwian) hieroglyphs, or with Egyptian hieroglyphs.[34] More remote possibilities are the Phoenician abjad or the Byblos syllabary.

Comparison of the disk's signs with those of Linear A inscriptions go back to Evans in 1909.[1] In1959, Benjamin Schwartz asserted a genetic relationship between the Phaistos Disc script and the Cretan linear scripts.[7]^: p.108 Similar claims were made also by Werner Nahm in 1975,[35] Torsten Timm in 2004,[36] and others.

Phaistos glyph 12

Phaistos glyph 43

Phaistos glyph 31

Some of these proposals point to simiarities between some glyphs, such as 12 (SHIELD), 43 (STRAINER), and 31 (EAGLE) to both Linear A and Linear B characters, and conjecture that they may have the same phonetic values — respectively 'qe', 'ta', and 'ku'. Based on the Linear A character distribution patterns collected by Giulio Facchetti,[37] Torsten Timm goes as far as identifying 20 of the 45 characters with Linear A/B signs.[38]

Phaistos glyph 01

Phaistos glyph 02

Phaistos glyph 11

Phaistos glyph 12

Phaistos glyph 39

Parallels between the Phaistos disk script and Anatolian hieroglyphs were proposed, among others, by S. Davis in 1961[39][40][41] and Jan Best and Fred Woudhuizen in 1988[42][43] In 2004, Winfried Achterberg and others proposed an extensive mapping to Anatolian hieroglyphs, which led them to a full decipherment claim.[5]

The following table summarizes the proposed identifications of Phaistos signs with Linear A/B, the Arkalochori Axe glyphs, and Luwian hieroglyphs:

No.	Sign	Linear A	Arkalochori Axe	Luwian hieroglyphs[5]
01				'SARU'
02			04,07,10	'A2'
10		AB79 'ZU'
11				'SOL SUUS'
12		AB78 'QE'		'TURPI'
15		A364 B232 ��
16		AB74 'ZE' ?
17		A322
18		AB37 'TI'
19		AB31 'SA'	11
22		A318
23		AB05 'TO' or AB06 'NA'	13
24		AB54 'WA'
25		AB86 [44]
29		AB80 'MA'	08
30		AB13 'ME', AB85?
31		AB81 'KU'
34		AB39 'PI'
35		AB04 'TE'	09
36		AB30 'NI'
39		AB28 'I'	02	'TARHUNT'
40		AB26 'RU' or AB27 'RE'
43		AB66 'TA2'
45		AB76 'RA2'

Decipherment claims can be categorized into linguistic decipherments, identifying the language of the inscription, and non-linguistic decipherments. A purely logographical reading is not linguistic in the strict sense: while it may reveal the meaning of the inscription, it will not allow for the identification of the underlying language.

Unless said otherwise, the attempts below assumed the right-to-left (clockwise, edge-to-center) reading direction, starting with side A.

• George Hempl, 1911: interpretation as Ionic Greek, syllabic writing.[45]
• Florence Stawell, 1911: interpretation as Homeric Greek, syllabic writing); B-side first.[46]
• Albert Cuny, 1914: interpretation as an ancient Egyptian document, syllabic-logographic writing.
• Benjamin Schwartz, 1959: interpretation as Mycenean Greek, syllabic writing, comparison to Linear B.[7][8]
• Jean Faucounau, 1975: interpretation as "proto-Ionic" Greek, syllabic writing.[47][48][49]
• Vladimir I. Georgiev, 1976: interpretation as Hittite language, syllabic writing; reading outward.
• Steven R. Fischer, 1988: interpretation as a Greek dialect, syllabic writing.[50]
• Kjell Aartun, 1992: interpretation as a Semitic language, syllabic writing; reading outward.[2]
• Derk Ohlenroth, 1996: interpretation as a Greek dialect, alphabetic writing; reading outward; numerous homophonic signs.[51][2]
• Adam Martin, 2000: interpretation as a Greek-Minoan bilingual text, alphabetic writing; reading outward, side A as Greek, side B as Minoan.[52]
• Achterberg et al., 2004: interpreted as Luwian.[5]
• Torsten Timm, 2005: syllabic writing, comparison to Linear A; B-side first.[38]
• Gareth Alun Owens, 2007: interpretation as Indo-European, syllabic writing, comparison to Linear A.[53]

• Paolo Ballotta, 1974: interpretation as logographic writing.
• Leon Pomerance, 1976: interpretation as astronomical document.
• Reiner J. van Meerten, 1977: interpretation as documentation of a gift to Minos.[54]
• Peter Aleff, 1982:interpretation as ancient gameboard.[55][56]
• Ole Hagen, 1988: interpretation as calendar.[57]
• Harald Haarmann, 1990: interpretation as logographic writing.
• Bernd Schomburg, 1997: calendar interpretation, logograms.[58]
• Patrick Berlingame, 2010: interpretation as the mythical labyrinth.
• Hermann Wenzel, 1998: astronomical interpretation.[59]
• Alan Butler, 1999: interpretation as calendar.
• Friedhelm Will, 2000: interpretation as number-philosophically-document of "Atlantean" origin.
• Axel Hausmann, 2002: document from Atlantis, dated to 4400 B.C., logographic reading.
• Helène Whittaker, 2005: a votive miniature version of a game board similar to the Egyptian Mehen.
• Wolfgang Reczko, 2009: eclipse table for Knossos, covering 4 Saros cycles of 18 years each.[60]

• Bennett, Emmett L. (1996) — Aegean Scripts, (in The World's Writing Systems, Peter T. Daniels and William Bright (Eds.) Oxford: University Press. ISBN 0-19-507993-0
• Chadwick, John. The Decipherment of Linear B, Cambridge University Press, 1958.
• Faure, P. "Tourne disque", l'énigme du disque de Phaistos, Notre Histoire n°213, October 2003 (PDF 0.7 Mb).
• Gaur, Albertine. 1984 — A History of Writing — Charles Scribner's Sons.
• Kober, Alice. The Minoan Scripts: Facts and Theory, American Journal of Archaeology, Vol. 52, No. 1 (1948), pp. 82–103.
• Sornig, Karl (2006). "The ultimate assessment". Grazer Linguistische Studien (65): 151–155.
• International Phaistos Disk Conference 2008, sponsored by Minerva Magazine. abstracts

• Aartun, Kjell, 'Der Diskos von Phaistos; Die beschriftete Bronzeaxt; Die Inschrift der Taragona-tafel' in Die minoische Schrift : Sprache und Texte vol. 1, Wiesbaden, Harrassowitz (1992) ISBN 3-447-03273-1
• Ephron, Henry D, (1962), "Tharso and Iaon: The Phaistos Disk, Harvard Studies in Classical Philology, Vol. 66. (1962), pp. 1–91. JSTOR URL
• Gordon, F. G. 1931. Through Basque to Minoan: transliterations and translations of the Minoan tablets. London: Oxford University Press.
• Hausmann, Axel, Der Diskus von Phaistos. Ein Dokument aus Atlantis, BoD GmbH (2002), ISBN 3-8311-4548-2.
• Thomas G. Palaima, Emmet L. Bennet, Jr., Michael G.F. Ventris, Alice E. Kober, "Cryptanalysis, Decipherment and the Phaistos Disc.", in M.-L. Nosch and H. Landenius-Enegren eds., Aegean Scripts, (Incunabula Graeca 105, Rome: 2017) vol. 2, pp. 771-788
• Polygiannakis, Ο Δισκος της Φαιστού Μιλάει Ελληνικά (The Phaistos disk speaks in Greek), Georgiadis, Athens (2000).
• Pomerance, Leon, The Phaistos Disk: An Interpretation of Astronomical Symbols, Paul Astroms forlag, Goteborg (1976). reviewed by D. H. Kelley in The Journal of Archeoastronomy (Vol II, number 3, Summer 1979)
• Whittaker, Helène (2005). "Social and Symbolic Aspects of Minoan writing". European Journal of Archaeology. 8 (1): 29–41. doi:10.1177/1461957105058207. S2CID 162881074.