[Note: All Basque words are in Italics and Bold-faced Green]
 
|    
  OGAM INSCRIPTIONS IN SCOTLAND * (Contact)                 In his book
  "The
  Symbol Stones of Scotland" (1984), Dr. Anthony Jackson,
  anthropologist at the University of Edinburgh, illustrated and transliterated
  more than thirty Ogam inscriptions found in Scotland and remarked that the
  best of efforts by linguists and others had not resulted in even one translation.
  There had been few problems transliterating them, but no one had been able to
  do anything with the "meaningless" series of letters obtained. In
  October following this work with an unpublished monograph called "Pictish Symbol
  Stones?" in which he updated his earlier research.
  Probably referring to efforts of Henri, Jackson wrote, "There is a
  popular theory that they are Basque but this does not work either" (p.118).
  Jackson also commented:             "It is
  curious that this small number of Ogam inscriptions has caused more headaches
  than all the other problems of the Picts put together. As one leading
  archaeologist put it: it is not really the fault of the Picts but the
  interpreters of the Picts that are to blame! (p. 117).             This remark was
  so true, but Jackson decided to give up entirely on translating the puzzling
  writings.  He wrote: "All
  research along linguistic lines has ground to a halt, unsurprisingly"
  (p.135) and: "It is clear that the Ogam inscriptions are numerically
  based and not linguistic" (p.153). 
  In other words he thought they were numerical magic, possibly a form
  of numerology, inscribed on the ancient standing stones to overcome the
  pre-Christian magic: "thus we seem to have a battle between rival
  magics" (p. 154). Edo Nyland agreed with his suggestion that magic
  is involved, because the inscriptions are so complicated in design that it is
  hard to believe that they were intended to be read by the common people or
  even by most of the clergy; they belonged to a very different level of
  theology.              In 1968 a Basque
  scholar from France, Henri Guiter, thought he could see Basque words
  in the transliterated inscriptions and tried to make sense of some of them.
  He published two papers in French, which received mixed reviews such as from Oliver  Padel who could not find the first paper, but "if
  one is to judge by the information supplied in the second, this is no great
  loss". Another person who criticized Guiter was Douglas Gifford,
  Dept. of Spanish of St. Andrew’s University in Scotland. In a 1969 radio
  talk, he said that Guiter had twisted the evidence, but also suggested that
  the Basque connection was worth a further look. Nyland then took this
  ‘further look’ and decided to include Guiter’s work in this article because
  his approach was so very different from anyone else’s. The reader will see
  that his translations appear to make little sense. The people who composed
  the inscriptions were a great deal more sophisticated linguistically and
  mathematically than our modern scholars have ever given them credit for.
  Guiter’s effort had also been published in Spanish in a booklet called
  "Garaldea" by Federico Krutwig and the
  Spanish translations of Guiter’s effort are shown here.               
  Dr. Gifford’s suggestion that Basque could well be the language of the
  Ogam inscriptions was supported by genetic and linguistic evidence in Ireland
  and Scotland. Geneticist Dr. Cavalli-Sforza from Stanford University had
  published a world map in Scientific American (Nov. 1991), showing the
  distribution of the Rh-negative people. The populations with the highest
  proportion of their members with Rh-negative blood
  were found among the Berbers in Morocco, the Basques in Euskadi, and the dark
  featured peoples of Northern Ireland and Scotland, all with over 25% of the
  people with this blood peculiarity. He commented "... the resulting
  pattern roughly coincides with anthropological reconstructions of ancient
  migrations." Of these four groups, only the Basques still spoke
  their pre-Christian language. It was therefore reasonable to suggest that the
  entire migration had spoken this language. This possibility was crying out
  for proof. Fortunately a very large number of early inscriptions on stone,
  silver, brass, bone etc. were available; over 600 in Ireland and some 40 in
  Scotland. None of these inscriptions had ever been translated with certainty.
  Transliteration from the Ogam script had not been a problem, but only an
  apparently meaningless series of letters, mostly consonants, had been
  obtained. However, as considerable time and effort must have gone into making
  these inscriptions.  Edo Nyland
  assumed that some system of decoding had to exist.             
  From the moment that Edo Nyland tackled the problem, it appeared
  likely that most of the vowels had been removed for some good reason, based
  on a certain pattern. After a great deal of experimentation, it was found that
  the basic pattern had to have been VCVCVCV etc. This letter-pattern looked
  strikingly like that of thousands of Basque words such as: "ohitura" (custom). But, Basque being an agglutinated
  language, this word in itself was composed of three other roots, ohi-itu-ura:
  ohi (habit) itungaitz (disagreeable) urratu (to break), meaning: "Break that disagreeable habit",
  creating a VCV-VCV-VCV pattern. In addition, the vowels on either side of the
  hyphens were always the same, completing the formula: VCV1-V1CV2-V2CV3-V3CV
  etc. Nyland called this the "vowel-interlocking" or "VCV Formula".
  Trial and error proved that this was indeed the formula used in every Ogam
  inscription examined to date, without exception. For more examples, see "The Saharan
  Language".             Searching for
  linguistic evidence of Basque in the family and geographical names of these
  countries, in Scotland many family names immediately stood out, e.g:              MacKenzie,
  kentze is the act of depriving, of taking away,
  to steal from, probably referring to territory. The MacKenzie tribe was
  therefore known by their neighbours as the people who had conquered or taken
  something that didn’t belong to them. 
 
 
 
 
 
     THE SYSTEM OF ENCODING AN OGAM
  INSCRIPTION   1).  In the sentence to be inscribed, use only
  those Basque words which start with vowel-consonant-vowel (VCV). 
 
 
       (see Jackson
  1993, pages 117 - 152). 
     1) Restore the original letters: V becomes B, C and Q become K. 
      were removed. In case
  of double vowels, an H has usually been removed. Keep in mind that every
  consonant represents a           word. 
 
      the words that form
  the appropriate sentence.   ABOUT THE TRANSLATION             In this section,
  all three interpretations by Guiter, Jackson and Nyland
  are brought together for each of the inscriptions .  Let the reader be the judge. The order in which the inscriptions
  are presented is taken from Jackson’s 1993 publication "Pictish
  Symbol Stones?” The transliteration used is also taken from Jackson
  because his interpretation is superior to any other efforts. Translating Ogam
  is certainly no exact science, it is only the best possible approximation. It
  may well be that some of the inscriptions were designed to be magical, yet
  when they were finally translated, most made good sense from the standpoint
  of evangelizing a "heathen" country. Two of the larger inscriptions,
  Brodie B and Golspie, in spite of several hours of work, have so far resisted
  the decoding process. Some like Altyre and Cille Barra describe natural
  disasters that do not refer to evangelism. Aboyne B and Altyre are grave
  markers. Strictly adhering to the vowel interlocking between the VCV roots
  is the key to decoding the inscriptions.   Map showing the
  location of the following inscriptions.   
     BRESSAY. A)
  CRROSCC- B) NAHHTVVDDADD - C) DATTRR -    Guiter:
  Basque reading: Berriz Enekoaren Kroska naiz Udak
  daragina. 
 
   Jackson:
  A) 28 7x4 75 5x5x3    Nyland:  
 
 
   B:
  NAHHTBBDDADDS 
 
   C:
  DATTRR 
 
   D:
  ANNBENNISES 
      
  The disciples, as
  well as the flock (Mark 14:50), in general weakness were mocking during that moment
  of tribulation (Mark 15:17-20)    E:
  MEOODDRROANN 
 
   BURRIAN. IDBMIRRHANNURRAC
  TEEVVCERROCCS   Guiter:       Basque
  reading: Don
  kuorari ańu(ti)ra dan kerroke. 
 
   Nyland:
  IDBMIRRHANNURRAKTEEBBKERROKKS 
 
   WHITENESS. VNDAR   Guiter:
  No reading.   Nyland: BNDAR. 
 
   LUNNASTING. A)
  ETTECUHETTS - B) AHEHHTTANNN - C) HCCVVEVV - D) NEHHTONN   Guiter:
  Basque reading: Etxekoez
  aiekoan nahigabe ba nengoen. 
 
   Jackson: A 36 6x6 140 7x5x2 
   Nyland: A: ETTEKUHETTS   et.     eta    etariko
          one of our group 
   B:
  AHEHHTTANNN   ahe     aihe     aiher        
  full of anger 
   C:
  HKKBBEBB   .h.     aha    ahal         
    I wish 
   D:
  NEHHTONN.   ne
       ne     nebarrebak      brothers and
  sisters 
   The
  place name Lunnasting itself is interesting:  
 
 
     ST. NINIANS. BESMEQQNANAMMOVVVEZ   Guiter:
  Basque reading:    
  Eneko ba nago bez. 
 
   Jackson: 54 6x9 172 43x4   Nyland: BESMEKKNANAMMOBBEZ   .be     be     bedeinkagarri   blessed
  one 
     BIRSAY.
  1) MBOLMVNORRALVRR - 2) BQIAB   Guiter: Basque reading: None.    Jackson: 43 prime 170 5x2x17   Nyland: Birsay 1) MBOLMVNORRALBRR   m.       ma    maisu
            teacher 
   Birsay
  2) BKIAB   .b.     be     bedeinkagarri   the
  blessed one 
   BUCKQUOY. ETMIQMSSALLC   Guiter:
  Basque reading: None.   Jackson: 36 6x6 135 5x3x3x3   Nyland: ETMIKMSSALLIK   et.     eti    etika
            ethics 
   CUNNINGSBURGH. 1)
  IRO - 2a) EHTECONMORS - 2b) DOVHDDRS - 3a) ETTECA - 3b) VDATTVB 3c) RTT   Guiter: Basque reading A few individual
  words only.   Jackson: (1): 12 3x440 5x2x4   Nyland: 1: IRO (on stone slab)   iro      iro    irol
            privy, outhouse 
   2a:
  EHTEKONMORS   eh.     ehu    ehun
             hundreds 
   2b:
  DOBHDDRS   do       do    doatsutasun    
  happiness 
   3a:
  ETTEKA   et.     eti    etikoa       
    ethical 
   3b:
  BDATTBB   .b.     abe    abegitasun     
  fondness of 
   3c:
  RTT   This
  last inscription has no identifiable vowel and therefore is not translatable with
  the vowel-interlocking method.      POOL. RVMVORC   Guiter: Basque reading: None.   Jackson: 23 prime 75 5x5x3   Nyland: Pool: RBMBORK   .r.     ara    arraro
          strange, odd 
   GURNESS.    NEITTEMTOS M0CS    Guiter: Basque
  reading: None.    Jackson: 50 5x10 189 7x3x3x3   Nyland: NEITTEMTOSMOKS.   .ne     ene    enekin       
  with me 
   NORTH UIST. H
  QUNCENTC T   Guiter:    Basque reading:
  Belaskuanuk..ta 
 
   Jackson: 37 prime 119 7x17   Nyland: HKUNKENTKT   .h.     ohi    ohitu
           to get used to 
     WEETING. 1)
  VLVEVVUTE - 2) GEDEVIM DOS   Guiter: Basque reading: None.   Jackson: 1) 28 7x4 84 7x3x2x2   Nyland: Weeting 1: Jackson: BLBEBBUTE, Nyland:
  BLBEBBETE    .b.     aba    abagadune
       occasion 
   Weeting
  2: GEDEBIMDOS   .ge     age    agerpen       
  revelation 
 
   BRANDSBUTT 8/45. IRATADDOARENS   Guiter:   
  Basque reading:
  Iratakoaren. 
 
   Jackson: 40 5x8 123 3x41   Nyland: IRATADDOARENS (two possible
  translations)   ira     ira    irakatsi     
  to preach a sermon 
   Nyland:  
  IRATADDOARENS 
 
     INCHYRA - A1:
  OOTTLIETRENOIDDORS 
 
 
 
   Guiter:   
  Basque reading: Etorkoaren
  ...holoi...ina otsa utz diet dinua? 
 
   Jackson: A1) 60 5x12 225 5x5x5x3    Nyland: A1: OOTTLIETRENOIDDORS   o.o     oho    ohoregabe
        dishonored 
   A2:
  UHTUOAGED   uh.     uhe    uherdura       
  confusion 
   B1:
  INEHHETESTIE.   ine     ine    inertzia       
  passive, downtrodden 
   B2:
  INNE –    in.     ine    inertzia       
  passive, downtrodden 
   C:
  SETU   .se     ase    asetu        
     to be filled with 
 
     KEISS 41/7: NEHTETRI   Guiter:   
  Basque reading: Nauke tagona. 
 
   Jackson: 30 5x6 95 5x19   Nyland: NEHTETRI   ne      ne     negarreztatu   
  grieving 
   LOGIE 8/5: CALTQUGuiter:    Basque reading: Kalkakoa 
 
   Jackson: 18 6x3 70 5x2x7   Nyland: KALTKU   .ka     aka    akabu        
    death 
     ABOYNE A: NEHHTVROBBACC
  – ENNEVV   ABOYNE B: MAQQOTALLUORR   Guiter:   
  Basque reading: Lemako
  da lurrpe. Dator doaken enea. 
 
   Jackson: A) 59 prime 149 prime   Nyland: A) NEHHTBROBBAKK – ENNEBB   .ne     ene    enetan
          always 
   en.     ene    enetan        
  always  
   B:
  MAKKOTALLUORR.   .ma     ama    ama
             mother 
     BRODIE A 31/5: VONECCO   BRODIE B 31/5: RAMINNGCHQODTOSLMBS   BRODIE C 31/5:
  EDDARRNONR TTI   Guiter:   
  Basque reading: Idarreko noa doa mokorra erala behar
  aikaz bedi.Du sutu  ocean iasoa
  lurreko                                                 
  karrak. Ba
  lo elhurra-be dago, haike, aikako ibaia du. 
 
   Jackson: A) 24 8x3 58 29x2   Nyland: A: BONEKKO.    .bo     abo    aboskatu
         to express 
   B:
  RAMINNGKHKODTOSLMBS.             This
  fairly long inscription is a complicated puzzle, which has not yet yielded
  its secret, probably because of the difficulties with reading the eroded
  inscription.   C:
  EDDARRNONR TTI   ed.     eda    edan         
     to drink 
   GOLSPIE 17/31: ALLHHALLORREDDMEOO
  – NUUVALHNRERR   Guiter:   
  Basque reading: Aldalurrekoak
  hartza lotu zuan. 
 
   Jackson: 88 8x11 320 5x8x8   Nyland: This large inscription looks authentic
  and should have given up its secrets, but I didn’t succeed yet in decoding
  it.     LATHERON 40E/41: DUVNODNNATMAOONAHATO   Guiter:   
  Basque reading: Doana da Eneko t’ekaitsua 
 
   Jackson: 56 8x7 208 13x4x4   Nyland: DUBNODNNATMAOONAHATO   .du     edu    eduki        
    to have 
   SCOONIE -/31: EDDARRNOSN   Guiter:   
  Basque reading: udara zan onsa. 
 
   Jackson: 35 5x7 106 53x2   Nyland: EDDARRNOSN   ed.     ede    eder egin
        to give pleasure 
   ALTYRE: AMMAQQHTALLMVBVMAA-HHRRASSUDDS   Guiter:
  Basque reading: None.   Jackson: 66 11x6 281 prime    Nyland: AMMAKKHTALLMBBBMAAHHRRASSUDDS   am.     ama    ama   
            mother 
   ABERNETHY: QMI.   Guiter:
  Basque reading: None.   Jackson: 11 prime 37 prime   Nyland: KMI   .k.     ika    
  ikara          fears 
   This is the first translation that appeared. KMI is very short,
  doesn’t leave much to work with, and cannot be translated with certainty.   AUQUHOLLIE: VUUNON - TEDOV – BB   Guiter:   
  Basque reading: Hila du ileko obiak. 
 
   Jackson: 37 prime 138 6x23    Nyland:
  VUUNON. The
  transliteration of this inscription is a problem. Only part makes
  sense, the VCV "uno" does not                 exist
  and "BB" has no vowel.   (TEDOV)
  should read: TSOLV   .te     ate    atedanbada   
     knock on the door 
   NEWTON: Jackson:
  A: IDDARQNNNVORRENNIEUA B: IOSRE 
   Guiter:   
  Basque reading: Idarkoari hor Eneko dio zagor. 
 
   Jackson: A) 77 7x11 210 7x3x10   Nyland: A:  IEARKNNNBDRRENNIEUA   i.e     ihe    ihesleku      
  find shelter 
   NEWTON B: IOSRZ   i.o     iho    ihortziri      
  thunder 
   DUNADD A: AESD - T - V - LVA – TV   DUNADD B: L---VIRRAMDNA   Guiter:
  Basque reading: None.   Jackson: A) 29 prime 96 4x4x6   Nyland: A: AESD - T - B - TB.   This
  inscription is too fractured to do anything with it.   B:
  L — BIRRAMDNAI   L ?            lagun?
           friends? 
   CILLE BARRA STONE   TIRTHURKIRTHUS;INRRISKURSSIARISTA:A             This
  stone was removed in 1865 from the Cille Barra cemetary (Isle of Barra) and
  taken to the Museum of Antiquities in Edinburgh. It was always thought to be
  a gravestone, which it obviously is not. The transliteration was copied from
  a local tourist pamphlet. The twin islands Barra-Vatersay are the most
  southerly populated islands in the Outer Hebrides of Scotland. The
  inscription is not found in Jackson’s or Guiter’s writings.   TIRTHURKIRTHUS   .ti     uti    utikan!
         get away from here! 
   INRRISKURSSIARISTA:A   in.     ino    inor
            everyone             Everyone
  is dismayed, petrified and overwhelmed to be eye-witness to this shocking
  flood from the beginning; to dry up we escaped to this leaky shelter.    CONCLUSION             It must be
  pointed out that these are not Pictish Ogams; instead, they are Irish
  Ogams in Pictland because they were written by early Irish evangelists who
  came to Scotland to convert the Pictish "heathens" to the Irish
  form of Christianity. All of the Irish and Scottish Ogam inscriptions that
  Edo Nyland has translated, and he has done almost one hundred, are written in
  the Basque language, without exception. Many, if not most, geographical and
  family names of Ireland and Scotland can also be translated with the Basque
  dictionary using the technique demonstrated above. Considering the evidence,
  it appears certain that prior to the coming of Roman Catholicism in about 650
  A.D., the Basque language, or an earlier form of it, was spoken as the
  popular language of the islands. This language was generally referred to
  by continental evangelists as the "Iron Language", also called Pictish
  in Scotland and Cruithin in Ireland. It seems to indicate
  that the Basque language hasn’t changed much over the past 1,500 years.     ILLUSTRATIONS:   Figure 1 Reprinted by permission from Anthony Jackson,
  ‘Pictish Symbol Stones ?’ 1993   
        Figure 2 Reprinted by permission from Anthony
  Jackson, ‘Pictish Symbol Stones ?’ 1993   
       Figure 3. Reprinted by permission from Anthony
  Jackson, ‘Pictish Symbol Stones ?’ 1993   
         Figure 4 Reprinted by permission from Anthony
  Jackson, ‘Pictish Symbol Stones ?’ 1993   
     Figure 5. Reprinted by permission from Anthony
  Jackson, ‘Pictish Symbol Stones ?’ 1993   
     | 
 
 
==========================================
For further detail, please
refer to:
 
          Nyland, Edo.  2001.  Linguistic Archaeology: AnIntroduction. Trafford Publ., Victoria, B.C., Canada.
               ISBN 1-55212-668-4. 541 p. [
see abstract & summary]
 
          Nyland, Edo.  2002. 
Odysseus and the Sea Peoples: A 
               Bronze Age History of Scotland  Trafford Publ., Victoria, 
               B.C., Canada. 
307 p.   [see
abstract & summary].
 
------------------------------------------------------------------------------------------------------------------------