Linear B
Linear B | |
---|---|
Script type | with additional Logograms |
Time period | Late Bronze Age |
Status | Extinct |
Direction | Left-to-right |
Languages | Mycenaean Greek |
Related scripts | |
Parent systems | Linear A
|
Sister systems | Cypro-Minoan syllabary |
ISO 15924 | |
ISO 15924 | Linb (401), Linear B |
Unicode | |
Unicode alias | Linear B |
U+10000–U+1007F syllabic signs U+10080–U+100FF logograms | |
Template:History of the Greek language
Linear B is a script that was used for writing Mycenaean, an early form of Greek. It predated the Greek alphabet by several centuries (ca 13th but perhaps as early as late 15th century BC) and seems to have died out with the fall of Mycenaean civilization. Most of the tablets inscribed in Linear B were found in Knossos, Cydonia,[1] Pylos, Thebes and Mycenae.[2] The succeeding period, known as the Greek Dark Ages, provides no evidence of the use of writing.
The script appears to be related to Linear A, an undeciphered earlier script used for writing the Minoan language, and the later Cypriot syllabary, which recorded Greek. Linear B consists of around 87 syllabic signs and a large repertory of semantographic signs. These “signifying” signs stand for objects or commodities, but do not have phonetic value and are never used as word signs in writing a sentence.
The application of Linear B was confined to administrative contexts. In all the thousands of tablets, a relatively small number of different "hands" have been detected: 45 in Pylos (west coast of the Peloponnese, in southern Greece) and 66 in Knossos (Crete).[3] From this fact it could be theorized that the script was used only by some sort of guild of professional scribes who served the central palaces. Once the palaces were destroyed, the script disappeared.
The script
Linear B has roughly 200 signs, divided into syllabic signs with phonetic values and logograms (or ideograms) with semantic values. The representations and naming of these signs has been standardized by a series of international colloquia starting with the first in Paris in 1956. After the third meeting in 1961 at the Wingspread conference center in Racine, Wisconsin, a standard proposed primarily by Emmett L. Bennett, Jr., became known as the Wingspread Convention, which was adopted by a new organization, CIPEM, affiliated in 1970 by the fifth colloquium with UNESCO. Colloquia continue: the 13th is scheduled for 2010 in Paris.[4]
Many of the signs are identical or similar to Linear A signs; however, Linear A, which encoded the unknown Minoan language, remains undeciphered and we cannot be sure that similar signs had similar values.[5]
Syllabic signs
The grid developed during decipherment by Ventris and Chadwick of phonetic values for syllabic signs is shown below.[6] Read the initial consonant in the left-most column and the vowel from the top row beneath the title. The transcription of the syllable (it may not have been pronounced that way) is listed next to the sign along with Bennett's identifying number for the sign preceded by an asterisk (as was Ventris' and Chadwick's convention).[7] In cases where the transcription of the sign remains in doubt, Bennett's number serves to identify the sign.[8] The signs on the tablets and other ancient artifacts often show considerable variation from each other and from the representations below. Discovery of the reasons for the variation and possible semantic differences is a topic of ongoing scholarship in Mycenaean studies.
Recognised signs of shape V, CV[9] | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|
-a | -e | -i | -o | -u | ||||||
𐀀 | a
*08 |
𐀁 | e
*38 |
𐀂 | i
*28 |
𐀃 | o
*61 |
𐀄 | u
*10 | |
d- | 𐀅 | da
*01 |
𐀆 | de
*45 |
𐀇 | di
*07 |
𐀈 | do
*14 |
𐀉 | du
*51 |
j- | 𐀊 | ja
*57 |
𐀋 | je
*46 |
𐀍 | jo
*36 |
||||
k- | 𐀏 | ka
*77 |
𐀐 | ke
*44 |
𐀐 | ki
*67 |
𐀒 | ko
*70 |
𐀓 | ku
*81 |
m- | 𐀔 | ma
*80 |
𐀕 | me
*13 |
𐀖 | mi
*73 |
𐀗 | mo
*15 |
𐀘 | mu
*23 |
n- | 𐀙 | na
*06 |
𐀚 | ne
*24 |
𐀛 | ni
*30 |
𐀜 | no
*52 |
𐀝 | nu
*55 |
p- | 𐀞 | pa
*03 |
𐀟 | pe
*72 |
𐀠 | pi
*39 |
𐀡 | po
*11 |
𐀢 | pu
*50 |
q- | 𐀣 | qa
*16 |
𐀤 | qe
*78 |
𐀥 | qi
*21 |
𐀦 | qo
*32 |
||
r- | 𐀨 | ra
*60 |
𐀩 | re
*27 |
𐀪 | ri
*53 |
𐀫 | ro
*02 |
𐀬 | ru
*26 |
s- | 𐀭 | sa
*31 |
𐀮 | se
*09 |
𐀯 | si
*41 |
𐀰 | so
*12 |
𐀱 | su
*58 |
t- | 𐀲 | ta
*59 |
𐀳 | te
*04 |
𐀴 | ti
*37 |
𐀵 | to
*05 |
𐀶 | tu
*69 |
w- | 𐀷 | wa
*54 |
𐀸 | we
*75 |
𐀹 | wi
*40 |
𐀺 | wo
*42 |
||
z- | 𐀼 | za
*17 |
𐀽 | ze
*74 |
𐀿 | zo
*20 |
Special and unknown signs
In addition to the grid the first edition of Documents contained a number of other signs termed "homophones" because they appeared at that time to resemble the sounds of other syllables and were transcribed accordingly: pa2 and pa3 were presumed to be homophonous to pa. Many of these were identified by the second edition and are shown in the "special values" below.[10] The second edition relates: "It may be taken as axiomatic that there are no true homophones." The unconfirmed identifications of *34 and *35 as ai2 and ai3 were removed. pa2 became qa.[11]
Other values remain unknown, mainly because of scarcity of evidence concerning them.[10][12] Note that *34 and *35 are mirror images of each other but whether this graphic relationship indicates a phonetic one remains unconfirmed.
In recent times CIPEM has inherited the former authority of Bennett and the Wingspread Convention in deciding what signs are to be considered "confirmed" and how all the various categories of signs shall be officially represented. In editions of Mycenaean texts, those signs whose value has not been confirmed by CIPEM are always transcribed as numbers preceded by an asterisk (e.g. *64). CIPEM also allocates the numerical identifiers, and until such allocation, new signs (or obscured or mutilated signs) are transcribed as a bullet-point enclosed in square brackets: [•].
Spelling and pronunciation
The signs are approximations: each may be used to represent a variety of distinct combinations of sounds, within rules and conventions. The grid presents a system of monosyllabic signs of the type V/CV. Clarifiction of the 14 or so special values tested the limits of the grid model but Chadwick in the end concluded that even with the ramifications the syllabic signs can unexceptionally be considered monosyllabic.[13]
Possible exceptions, Chadwick goes on to explain, include the two diphthongs, ai and au, as in Ai-ku-pi-ti-jo for Aiguptios (Αἰγὐπτιος) and Au-ke-wa for Augewās (Αὐγείας).[14] However a diphthong is by definition two vowels united into a single sound and therefore might be typed as just V. Thus rai, as in e-rai-wo for elaiwon, is of the type CV. Diphthongs are otherwise treated as two monosyllables: a-ro-u-ra for aroura, of the types CV and V.[15] Lengths of vowels and accents are not marked.
Twe, two, dwe, dwo, nwa and the more doubtful swi and swa may be regarded as beginning with labialized consonants, rather than two consonants, even though they may alternate with a two-sign form: O-da-twe-ta and O-da-tu-we-ta for Odatwenta; A-si-wi-jo and A-swi-jo for Aswios. Similarly, rya, ryo and tya begin with palatalized consonants rather than two consonants: -ti-ri-ja or -ti-rya for -trya (later -tria).
The one sign Chadwick tags as the exception to the monosyllabic rule is pte, but this he attributes to a development pte<*pje as in kleptei<*klep-yei.
Linear B does not consistently distinguish between voiced and unvoiced stops (except in the dental series) and between aspirated and unaspirated stops even when these distinctions are phonemic in Mycenaean Greek. For example,[16] pa-te is patēr, pa-si is phāsi (p on the other hand never represents b: basileus is qa-si-re-u); ko-ro is korus, ka-ra-we is grāwes, ko-no is skhoinos. Exceptionally, however, the dentals are represented by a t-series and a d-series for unvoiced and voiced: to-so for tosos but do-ra for dōra; however, to-ra-ke for thōrākes. In other cases aspiration can be marked but is optional: pu-te for phutēr ("planter"), but phu-te-re for phutēres ("planters"). Initial aspiration may be marked only in the case of initial a and rarely: ha-te-ro for hateron (later heteron),[17] and yet a-ni-ja for hāniai.
The j-series represents the semivowel equivalent to English "y", and is used word-initially and as an intervocalic glide after a syllable ending in i: -a-jo for -αῖος (-aios); a-te-mi-ti-jo for Ἀρτεμιτιος (Artemitios). The w-series similarly are semivowels used word-initially and intervocalically after a syllable ending in u: ku-wa-no for kuanos.[18]
The r-series includes both the /r/ and /l/ phonemes.
The "q"-series is used for the labiovelars resulting from Proto-Indo-European /kʷ/, /gʷ/, and /gʷʰ/, the latter being devoiced to /kʷʰ/ in Proto-Greek.
Logograms ("ideograms")
Linear B also uses a large number of logograms,[citation needed] which typically go by the old term "ideogram" in the literature on Linear B. They express:
- The type of object concerned (e.g. a cow, wool, a spear)
- A unit of measure.
They are typically at the end of a line before a number and appear to signify the object to which the number applies. Many of the values remain unknown or disputed. Some commodities such as cloth and containers are divided into many different categories represented by distinct ideograms. Livestock may be marked with respect to their sex.
The numerical references for the ideograms were originally devised by Ventris and Bennett, divided into functional groups corresponding to the breakdown of Bennett's index. These groups are numbered beginning 100, 110, 120 etc., with some provision of spare numbers for future additions; the official CIPEM numberings used today are based on Ventris and Bennett's numbering, with the provision that three or four letter codes (written in small capitals), based on Latin words that seemed relevant at the time, are used where the meanings are known and agreed. Unicode (as of version 5.0) encodes 123 Linear B ideograms.
The ideograms are symbols, not pictures of the objects in question - e.g. one tablet records a tripod with missing legs, but the ideogram used is of a tripod with three legs. In modern transcriptions of Linear B tablets it is typically convenient to represent an ideogram by its Latin or English name or by an abbreviation of the Latin name. Ventris and Chadwick generally used English; Bennett, Latin. Neither the English nor the Latin can be relied upon as an accurate name of the object; in fact, the identification of some of the more obscure objects is a matter of exegesis.
It has been suggested that this section be split out into another article titled List of Linear B ideograms. (Discuss) (January 2009) |
Glyph | Codepoint[19] | Bennett[20] | CIPEM[21] | English[22] |
---|---|---|---|---|
People and Animals | ||||
U+10080 | 100[23] A- | VIR vir |
MAN | |
U+10081 | 102 A- | MUL mulier |
WOMAN | |
U+10082 | 104 Cn | CERV cervus |
DEER | |
𐂃 | U+10083 | 105 Ca S- | EQU equus |
HORSE |
𐂄 | U+10084 | 105 Ca | EQUf | mare[24] |
𐂅 | U+10085 | 105 Ca | EQUm | stallion |
𐀥 | U+10025 | 106 QI *21 |
OVIS ovis |
SHEEP |
𐀥 | WE *75 |
we-ka-ta Bous ergatēs |
"Adjunct to ox" (1973)[25] | |
𐂆 | U+10086 | 106b C- D- | OVISf | EWE |
𐂇 | U+10087 | 106a C- D- | OVISm | RAM |
𐁒 | U+10052 | 107 RA *22 |
CAP capra |
GOAT |
𐂈 | U+10088 | 107b C- Mc | CAPf | SHE-GOAT |
𐂉 | U+10089 | 107a C- | CAPm | HE-GOAT |
𐁂 | U+10042 | 108 AU *85 C- |
SUS sus |
PIG |
𐂊 | U+1008A | 108b C- | SUSf | SOW |
𐂋 | U+1008B | 108a C- | SUSm | BOAR |
𐀘 | U+10018 | 109 MU *23 C- |
BOS bos |
OX |
𐂌 | U+1008C | 109b C- | BOSf | COW |
𐂍 | U+1008D | 109a C- | BOSm | OX/BULL |
Units of Measurement | ||||
110 | Z kotylai |
Volume Cup[26] | ||
111 | V khoinikes |
Volume | ||
112 | T | Dry | ||
113 | S | Liquid | ||
114 | Weight | |||
*21 | Weight | |||
*2 | Weight | |||
115 | P | Weight | ||
116 | N | Weight | ||
117 | M dimnaion[27] |
Weight | ||
118 | L talanton |
TALENT | ||
*72 G- | Bunch? | |||
*74 S- | Pair | |||
*15 S- | Single | |||
*61 | Deficit | |||
By Dry Measure | ||||
𐂎 | U+1008E | 120 E- F- | GRA granum |
WHEAT |
𐂏 | U+1008F | 121 F- | HORD hordeum |
BARLEY |
𐂐 | U+10090 | 122 F- U- | OLIV oliva |
OLIVES |
𐂑 | U+1001B | NI *30 F |
FICUS | FIGS |
𐂑 | U+1000E | *65 | FARINA | FLOUR "some kind of grain"[28] |
𐂑 | U+10091 | 123 G- Un | AROM aroma |
CONDIMENT |
KO *70 G- |
Coriander | |||
U+1002D | SA *31 G- |
Sesame | ||
KU *81 G- |
Cumin | |||
SE *9 G- |
Celery | |||
MA *80 G- |
Fennel | |||
124 G- | PYC | cyperus | ||
𐂒 | U+10092 | 125 F- | CYP | cyperus? |
126 F- | CYP+KU | cyperus+ku | ||
𐂓 | U+10093 | 127 Un | KAPO | fruit? |
𐂔 | U+10094 | 128 G- | KANAKO | safflower |
By liquid measure | ||||
By weight | ||||
By weight or in units | ||||
Counted in units | ||||
Vessels[29] | ||||
𐃆 | U+100DF | 200 | sartago |
BOILING PAN |
𐃆 | U+100E0 | 201 | TRI tripus |
TRIPOD CAULDRON |
𐃆 | U+100E1 | 202 | poculum |
GOBLET? |
𐃆 | U+100E2 | 203 | urceus |
WINE JAR? |
𐃆 | U+100E3 | 204 Ta | hirnea |
EWER |
𐃆 | U+100E4 | 205 K Tn | hirnula |
JUG |
𐃆 | U+100E5 | 206 | HYD hydria |
HYDRIA |
𐃆 | U+100E6 | 207 | TRIPOD AMPHORA | |
𐃆 | U+100E7 | 208 | PAT patera |
BOWL |
𐃆 | U+100E8 | 209 | AMPH amphora |
AMPHORA |
𐃆 | U+100E9 | 210 | STIRRIP JAR | |
𐃆 | U+100EA | 211 | WATER BOWL? | |
𐃆 | U+100EB | 212 | SIT situla |
WATER JAR? |
𐃆 | U+100EC | 213 | LANX lanx |
COOKING BOWL |
Furniture | ||||
𐃆 | U+100C4 | 220 Ta | scamnum |
FOOTSTOOL |
𐃆 | U+100C5 | 225 | ALV alveus |
|
Weapons | ||||
𐃆 | U+100C6 | 230 R | HAS hasta |
SPEAR |
𐃇 | U+100C7 | 231 R | SAG sagitta |
ARROW |
𐃈 | U+100C8 | 232 Ta | *232 | ? |
𐃉 | U+100C9 | 233 Ra | DAGGER | |
𐃉 | U+100CA | 234 | GLA gladius |
SWORD |
Chariots | ||||
𐃌 | U+100CC | 240 Sc | BIG biga |
WHEELED CHARIOT |
𐃍 | U+100CD | 241 Sd Se | CUR currus |
WHEEL-LESS CHARIOT |
𐃎 | U+100CE | 242 Sf Sg | CAPS capsus |
CHARIOT FRAME |
𐃏 | U+100CF | 243 Sa So | ROTA rota |
WHEEL |
Archives
Corpus
The tablets are classified by the location of their excavation.
- KN Knossos: ca. 4360 tablets (not counting finds of Linear A).
- PY Pylos : 1087 tablets.
- TH Thebes: 99 tablets + 238 published in 2002 (L. Godart and A. Sacconi, 2002; see under Thebes tablets)
- MY Mycenae: 73 tablets.
- TI Tiryns: 27 tablets.
- KH Chania: 4 tablets.
- another 170 inscriptions in Linear B were found on vessels.
If it is genuine, the Kafkania pebble, dated to the 17th century BC, would be the oldest known Mycenean inscription, and hence the earliest preserved testimony of the Greek language.
Chronology
The main archives for Linear B are associated with these stages of Late Minoan and Helladic pottery:[30]
LM II (1425-1390 BC):
- Knossos, Room of the Chariot Tablets.
LM IIIA2 (1370-1340 BC) or IIIB (1340-1190 BC):
- Knossos, main archive.
LM IIIB:
- Chania, tablet Sq 1, 6659, KH 3 (possibly Linear B).
LH/LM IIIB1 end:[31]
- Chania, tablets Ar 3, Gq 5, X 6.
- Mycenae, tablets from Oil Merchant group of houses.
- Thebes, Ug tablets and Wu sealings.
LH IIIB2, end:
- Mycenae, tablets from the Citadel.
- Tiryns, all tablets.
- Thebes, Of tablets and new Pelopidou Street deposit.
- Pylos, all but five tablets.
Controversy
The Knossos archive was dated by Sir Arthur Evans to the destruction by conflagration at about 1400 BC (which baked and preserved the clay tablets), in the Late Minoan II (LM II) period. Evans made a career of excavating the Knossos site in the late 19th and early 20th centuries and creating the concept of Minoan civilization, which he believed was historically accurate.
This view stood until Carl Blegen excavated the site of ancient Pylos in 1939 and uncovered tablets inscribed in Linear B, one of the two scripts discovered at Knossos and named by Evans. Those tablets were fired in the conflagration that destroyed Pylos about 1200 BC, at the end of Late Helladic IIIB (LHIIIB). With the decipherment of Linear B by Michael Ventris in 1952, serious questions about Evans' date began to be considered. Most notably, Blegen said that the inscribed stirrup jars (an oil flask with stirrup-shaped handles) imported from Crete around 1200 were of the same type as those dated by Evans to the destruction of 1400. Blegen found a number of similarities between 1200 BC Pylos and 1400 BC Knossos and suggested the Knossian evidence be reexamined, as he was sure of the 1200 Pylian date.
The examination uncovered a number of difficulties. The Knossos tablets had been found at various locations in the palace and Evans had not kept exact records. Recourse was had to the day books of Evans' assistant, Duncan Mackenzie, who had conducted the day-to-day excavations. There were discrepancies between the notes in the day books and Evans' excavation reports. Worse, the two men had quarreled over the location and strata of the tablets, Mackenzie had called Evans a liar, and Evans had not only sacked him but made sure he did not excavate anywhere else.
The results of the reinvestigation were eventually published in a definitive work: Palmer, L.R. (1963). On the Knossos Tablets. Clarendon Press. {{cite book}}
: Unknown parameter |city=
ignored (|location=
suggested) (help); Unknown parameter |coauthors=
ignored (|author=
suggested) (help) It consists of two works, Leonard Palmer's The Find-Places of the Knossos Tablets and John Boardman's The Date of the Knossos Tablets. In this book Palmer plays the role, so to speak, of prosecuting attorney and Boardman of defending attorney; consequently, the dispute was known for a time as "the Palmer-Boardman dispute".
Like questions concerning the veracity of Heinrich Schliemann, the controversy soon escalated beyond the evidence, which set the world of classical scholarship looking for a way to resolve the question once and for all, a still unfulfilled hope. There appeared to be no "smoking gun" of Evans' mendacity; that is, he could in his excavation reports have simply been generalizing to resolve contradictions in the data. Moreover Blegen's arguments depended more on a preponderance of evidence rather than any single incontrovertible proof. No such incontrovertible proof has ever been found.
The real issue is whether sufficient reason exists to question Evans, since there is as much evidence for 1400 as there is for 1200. Without a solid reason to doubt Evans, the community of classical scholars tends to support a date of 1400 by default; that is, LM IIIA:2. As for LH IIIB, it likely begins in the 1310s or 1300s BC, after Mursilis II's sack of Miletus; it ends around 1200 BC.
The colours of the scholars can be identified by the dates they give for the tablets. This article utilizes an outline developed by Cynthia Shelmerdine, who is in the Boardman camp. While stating two possibilities for the main archive of Knossian tablets, she accepts a 1400 date for the Room of the Chariot Tablets. There still might have been two conflagrations and tablet firings, one in 1200 and one in 1400. As an example of a scholar who is in the Palmer camp, see Rutter, Jeremy B. (2000). "Lesson 25: The Linear B Tablets and Mycenaean Social, Political, and Economic Organization". Prehistoric Archaeology of the Aegean. Retrieved 2008-03-28. Rutter relies on a similarity of scribal hand between one of the Chania tablets and the Knossos tablets and dates all the tablets from 1350-1300 to 1200 BC.
Nearly every scholar presents their view as the generally accepted view or the one most proved by recent evidence. Regardless of how they may present their perceptions, the issue is very much open and the search for evidence continues.
Contents
The major cities and palaces used Linear B for records of disbursements of goods. Wool, sheep, and grain were some common items, often given to groups of religious people and also to groups of "men watching the coastline."
The tablets were kept in groups in baskets on shelves, judging by impressions left in the clay from the weaving of the baskets. When the buildings in which they were housed were destroyed by fires, many of the tablets were then fired.
Decipherment
The first clay tablet at Knossos was discovered by the British archaeologist Arthur Evans on March 31, 1900 and on April 6 he discovered a significant hoard of tablets (measuring 5x10 inches).
The convention for numbering the symbols still in use today was first devised by United States Professor Emmett L. Bennett, Jr., who, by 1950, had deciphered the metrical system. He was also an early proponent of the idea that Linear A and B represented different languages.
About the same time, Alice Kober studied Linear B and managed to construct grids, linking symbols that seemed to have a strong grammatical relationship. Kober noticed that a number of Linear B words had common roots and suffixes. This led her to believe that Linear B represented an inflected language, with nouns changing their endings depending on their case. However, some characters in the middle of the words seemed to correspond with neither a root nor a suffix. Because this effect was found in other, known languages, Kober surmised that the odd characters were bridging syllables, with the beginning of the syllable belonging to the root and the end belonging to the suffix. This was a reasonable assumption, since Linear B had far too many characters to be considered alphabetic and far too few characters to be logographic; therefore, each character should represent a syllable.
Using the knowledge that certain characters shared the same beginning or ending sounds, Kober built a table similar to the one above; her untimely death at age 43 in 1950 prevented her from possibly taking the final step or see others do it, namely to link the characters to actual phonetics.
Michael Ventris and John Chadwick performed the bulk of their decipherment of Linear B between 1951 and 1953. At first, Ventris chose his own numbering system, and agreed with Evans' hypothesis that Linear B was not Greek; however he later switched back to Bennett's system.
Based on Kober's work, and after making some inspired assumptions, Ventris was able to deduce the pronunciation of the syllables. To the amazement of Ventris himself, the deciphering of Linear B proved that it was a written form of Greek, in direct contradiction to the general scientific views of the times. Chadwick, an expert in historical Greek, helped Ventris decipher the text and rebuild the vocabulary and grammar of Mycenaean Greek.
Ventris' discovery was of immense significance, because it demonstrated a Greek-speaking Minoan-Mycenaean culture on Crete, and presented Greek in writing some 600 years earlier than what was thought at the time.
Interestingly, in 1935, the British School at Athens was celebrating its fiftieth anniversary with an exhibition at Burlington House, London. Among the speakers was Sir Arthur Evans, then in his eighty-fourth year and the teenager Michael Ventris was present in the audience.[32]
Unicode
Linear B is assigned Unicode range 10000–1007F for syllabograms and 10080–100FF for ideograms.
Linear B Syllabary[1][2] Official Unicode Consortium code chart (PDF) | ||||||||||||||||
0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | A | B | C | D | E | F | |
U+1000x | 𐀀 | 𐀁 | 𐀂 | 𐀃 | 𐀄 | 𐀅 | 𐀆 | 𐀇 | 𐀈 | 𐀉 | 𐀊 | 𐀋 | 𐀍 | 𐀎 | 𐀏 | |
U+1001x | 𐀐 | 𐀑 | 𐀒 | 𐀓 | 𐀔 | 𐀕 | 𐀖 | 𐀗 | 𐀘 | 𐀙 | 𐀚 | 𐀛 | 𐀜 | 𐀝 | 𐀞 | 𐀟 |
U+1002x | 𐀠 | 𐀡 | 𐀢 | 𐀣 | 𐀤 | 𐀥 | 𐀦 | 𐀨 | 𐀩 | 𐀪 | 𐀫 | 𐀬 | 𐀭 | 𐀮 | 𐀯 | |
U+1003x | 𐀰 | 𐀱 | 𐀲 | 𐀳 | 𐀴 | 𐀵 | 𐀶 | 𐀷 | 𐀸 | 𐀹 | 𐀺 | 𐀼 | 𐀽 | 𐀿 | ||
U+1004x | 𐁀 | 𐁁 | 𐁂 | 𐁃 | 𐁄 | 𐁅 | 𐁆 | 𐁇 | 𐁈 | 𐁉 | 𐁊 | 𐁋 | 𐁌 | 𐁍 | ||
U+1005x | 𐁐 | 𐁑 | 𐁒 | 𐁓 | 𐁔 | 𐁕 | 𐁖 | 𐁗 | 𐁘 | 𐁙 | 𐁚 | 𐁛 | 𐁜 | 𐁝 | ||
U+1006x | ||||||||||||||||
U+1007x | ||||||||||||||||
Notes |
Linear B Ideograms[1][2] Official Unicode Consortium code chart (PDF) | ||||||||||||||||
0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | A | B | C | D | E | F | |
U+1008x | 𐂀 | 𐂁 | 𐂂 | 𐂃 | 𐂄 | 𐂅 | 𐂆 | 𐂇 | 𐂈 | 𐂉 | 𐂊 | 𐂋 | 𐂌 | 𐂍 | 𐂎 | 𐂏 |
U+1009x | 𐂐 | 𐂑 | 𐂒 | 𐂓 | 𐂔 | 𐂕 | 𐂖 | 𐂗 | 𐂘 | 𐂙 | 𐂚 | 𐂛 | 𐂜 | 𐂝 | 𐂞 | 𐂟 |
U+100Ax | 𐂠 | 𐂡 | 𐂢 | 𐂣 | 𐂤 | 𐂥 | 𐂦 | 𐂧 | 𐂨 | 𐂩 | 𐂪 | 𐂫 | 𐂬 | 𐂭 | 𐂮 | 𐂯 |
U+100Bx | 𐂰 | 𐂱 | 𐂲 | 𐂳 | 𐂴 | 𐂵 | 𐂶 | 𐂷 | 𐂸 | 𐂹 | 𐂺 | 𐂻 | 𐂼 | 𐂽 | 𐂾 | 𐂿 |
U+100Cx | 𐃀 | 𐃁 | 𐃂 | 𐃃 | 𐃄 | 𐃅 | 𐃆 | 𐃇 | 𐃈 | 𐃉 | 𐃊 | 𐃋 | 𐃌 | 𐃍 | 𐃎 | 𐃏 |
U+100Dx | 𐃐 | 𐃑 | 𐃒 | 𐃓 | 𐃔 | 𐃕 | 𐃖 | 𐃗 | 𐃘 | 𐃙 | 𐃚 | 𐃛 | 𐃜 | 𐃝 | 𐃞 | 𐃟 |
U+100Ex | 𐃠 | 𐃡 | 𐃢 | 𐃣 | 𐃤 | 𐃥 | 𐃦 | 𐃧 | 𐃨 | 𐃩 | 𐃪 | 𐃫 | 𐃬 | 𐃭 | 𐃮 | 𐃯 |
U+100Fx | 𐃰 | 𐃱 | 𐃲 | 𐃳 | 𐃴 | 𐃵 | 𐃶 | 𐃷 | 𐃸 | 𐃹 | 𐃺 | |||||
Notes |
Notes
- ^ Hogan, C. Michael (2008). "Cydonia" (html). The Modern Antiquarian. Julian Cope.
{{cite web}}
: Unknown parameter|accesdate=
ignored (|access-date=
suggested) (help) - ^ Linnea Holmer Wren, David J. Wren and Janine M. Carter (1994) Perspectives on Western Art, Westview Press ISBN 0064301540
- ^ Hooker, J.T. (1980). Linear B: An Introduction. Bristol Classical Press UK. ISBN 0906515696.
- ^ Palaima, T.G. "A Brief History of CIPEM" (html). Comitē International Permanent des Ētudes Mycéniennes. Retrieved 2008-03-28.
{{cite web}}
: Text "work" ignored (help) - ^ Ventris and Chadwick (1973), page 37, quotes Bennett: "where the same sign is used in both Linear A and B there is no guarantee that the same value is assigned to it."
- ^ Ventris and Chadwick (1973), Fig. 4 on page 23 states the "Proposed values of the Mycenaean syllabary", which is mainly the same as the table included in this article. The "grid" from which it came, which was built up in "successive stages", is shown in Fig. 3 on page 20.
- ^ In the Unicode character names, Bennett's number has been rendered into a three-digit code by padding with initial zeros and preceding with a B (for "Linear B").
- ^ Ventris and Chadwick (1973), Fig. 9 on page 41 states Bennett's numbers from 1 through 87 opposite the signs being numbered. The table includes variants from Knossos, Pylos, Mycenae and Thebes opposite the same numbers.
- ^ In linguistics C and V in this type of context stand for consonant and vowel.
- ^ a b Ventris and Chadwick (1973), page 385.
- ^ Ventris and Chadwick (1973), pages 391-392.
- ^ Sign *89 is not listed in Ventris & Chadwick's (1973) tables but it does appear in the appendix of Bennett (1964) as part of the Wingspread convention.
- ^ Ventris & Chadwick (1973), pages 385-391.
- ^ Ventris and Chadwick use Roman characters for the reconstructed Mycenaean Greek and give the closest later literary word in Greek characters. Often the phonetics are the same but equally as often the reconstructed words represents an earlier form. Here the classical Greek was formed by dropping the w and lengthening the e to ei.
- ^ Ventris and Chadwick (1973), page 43.
- ^ The examples in this section except where otherwise noted come from the Mycenaean Glossary of Ventris & Chadwick (1973).
- ^ Ventris & Chadwick (1973), pages 388-391.
- ^ Ventris & Chadwick (1973), page 44.
- ^ Note that the codes do not represent all the glyphs, only the major ones.
- ^ This table follows the numbering scheme worked out by Ventris and Bennett and presented in Ventris and Chadwick (1973) in the table of Figure 10, pages 50-51. The superscript a refers to Bennett's "Editio a", "a hand from Pylos, of Class III." The superscript b refers to Bennett's "Editio b", "a hand of Knosses." The superscript c refers to Bennett's "Editio c", "a hand of Pylos, of Class I." The non-superscript letters represent the class of tablets, which precedes the individual tablet number; for example, Sa 787 is Tablet Number 787 of the class Sa, which concerns chariots and features the WHEEL ideogram.
- ^ Figure 10 in Ventris and Chadwick (1973) states only the English names of the logograms where they exist but the Latin is given where it exists in Bennett, Jr. Editor, Emmett L. (1964). Mycenaean Studies: Proceedings of the Third International Colloquium for Mycenaean Studies Held at "Wingspread," 4-8 September 1961. University of Wisconsin Press. pp. 258–259, "Ideogrammatum Scripturae Mycenaeae Transcriptio".
{{cite book}}
:|last=
has generic name (help); Unknown parameter|city=
ignored (|location=
suggested) (help) The "m" and "f" superscript are male and female. - ^ Given in capital letters if it repeats Ventris and Chadwick (1973) Figure 10; otherwise, in lowercase. Note that not all the CIPEM glyphs appear in Figure 10.
- ^ Ventris and Chadwick (1973) page 391: "100 MAN is now used for all forms of the ideogram, so that 101 and 103 are now suppressed."
- ^ Ventris & Chadwick either edition do not follow the Wingspread Convention here but have 105a as a HE-ASS and 105c as a FOAL.
- ^ The 1956 edition has "Kind of sheep"
- ^ Chadwick (1976) page 105.
- ^ "Double mina", Chadwick (1976) page 102.
- ^ Ventris & Chadwick (1973) page 392.
- ^ Ventris and Chadwick (1973) page 324 has a separate table.
- ^ This table is heavily indebted to Shelmerdine, Cynthia. "Where Do We Go From Here? And How Can the Linear B Tablets Help Us Get There?" (PDF). Retrieved 2008-03-27.
- ^ LM III is equivalent to LH III from a chronological perspective.
- ^ http://www.mansfield.edu/~art/papyrus1_stephanie_stambaugh.htm
Further reading
- Chadwick, John (1976). The Mycenaean World. Cambridge UP. ISBN 0-521-29037-6.
- Chadwick, John (1958). The Decipherment of Linear B. Second edition (1990). Cambridge UP. ISBN 0-521-39830-4.
- Chadwick, John (1987). Linear B and Related Scripts; "Reading the Past". Third impression (1997). University of California Press/British Museum. ISBN 0-520-06019-9. has the Enkomi clay tablet, circa 1500 BCE., examples of Linear B tablets, and translated, the basic Linear B syllabary, the Cypriot syllabary and discussions thereof, and short sections on Linear A, and the Phaistos Disk.
- Levin, Saul (1964). The Linear B Decipherment Controversy Re-examined. State University of New York Press. OCLC 288842.
- Robinson, Andrew (1995). The Story of Writing. Paperback edition (1999). Thames and Hudson. ISBN 0-500-28156-4. Chapter 6, Linear B, pp 108-119: discusses Arthur Evans, his work, the Cypriot clues, the syllabary, Alice Kober, the "Grid", and a sample tablet transliterated, and translated into English.
- Singh, Simon (2000). The Code Book. Anchor. ISBN 0-385-49532-3. for a general outline of the Linear B deciphering story, from Schliemman to Chadwick.
- Ventris, Michael (1988). Work notes on Minoan language research and other unedited papers. Edizioni dell'Ateneo 1988 Roma.
- Ventris, Michael and Chadwick, John (1956). Documents in Mycenaean Greek. Second edition (1973). Cambridge UP. ISBN 0-521-08558-6.
{{cite book}}
: CS1 maint: multiple names: authors list (link)
See also
External links
- Ager, Simon (1998–2009). "Linear B" (html). Omniglot. Retrieved 2009-01-06.
{{cite web}}
: CS1 maint: date format (link) - Clark, Curtis (1992–1998). "TrueType Fonts: Ancient alphabets and mythic symbols: Linear B". Curtis Clark.
{{cite web}}
: CS1 maint: date format (link) - Gavalas, Markos. "MYCENAEAN (Linear B) - ENGLISH Dictionary" (PDF). explorecrete.com. Retrieved 2009-01-11.
- Template:Dmoz
- Lo, Lawrence (1996–2005). "Linear B" (html). AncientScripts.com. Retrieved 2009-01-05.
{{cite web}}
: CS1 maint: date format (link) - McCreedy, David. "Gallery of Unicode Fonts: Linear B Syllabary" (html). WAZU, Japan. Retrieved 2009-01-11.
{{cite web}}
: Unknown parameter|coauthors=
ignored (|author=
suggested) (help) - Owens, Dr. Gareth (2005–2008). "Daidalika - Scripts and Languages of Minoan and Mycenaean Crete" (html) (in English and Modern Greek). Technological Educational Institute (TEI) of Crete. Retrieved 2009-01-09.
{{cite web}}
: CS1 maint: date format (link) CS1 maint: unrecognized language (link). - Palmer, Dr. Michael M. "The Linear B Syllabary" (html). Chapel Hill, NC: Greek-Language.com. Retrieved 2009-01-05.
- Rutter, Jeremy B (1996–1997). "The Prehistoric Archaeology of the Aegean" (html). Hanover, NH: The Foundation of the Hellenic World, Dartmouth College. Retrieved 2009-01-05.
{{cite web}}
: CS1 maint: date format (link) - Rutter, Jeremy B (1996–1997). "The Linear B Tablets and Mycenaean Organization" (html). Hanover, NH: The Foundation of the Hellenic World, Dartmouth College. Retrieved 2009-01-05.
{{cite web}}
: CS1 maint: date format (link) - Unicode code pages for the Linear B syllabary and logograms, including sample glyphs.