My God, my God, why have you forsaken me?

Jesus on the Cross

Most churches do a Passion Play this time of year, re-enacting the final moments of Jesus up to and including the crucifixion. Most of these Passion Plays tend to include Jesus’ final words as recorded in Matthew and Luke which appear in most Bibles transliterated as:

“Eloi! Eloi! Lama sabachthani?”

“How the heck do you pronounce *that*?” I am asked often enough. “Eh-loy eh-loy llama sab-ach!-thane-y?”

And my answer is: You don’t.

In truth, this phrase has been subject to a game of telephone, which started in Aramaic and twisted its way through Greek, and some German spelling conventions, before landing in English.

This phrase is an Aramaic translation of the beginning of Psalm 22, “My God, my God, why hast thou forsaken me? Why art thou so far from helping me, from the words of my groaning?”

As we can see from extant translations in other Aramaic dialects, in Jesus’ native Galilean Aramaic, it was most likely rendered:

אלהי אלהי למה שבקתני
əlahí əlahí ləmáh šəvaqtáni

(The funny upside-down e signifies a shewa, a vowel kind of like the a in “above.”)

When the Gospel writers were compiling their work in Greek, they ran into some interesting problems. Mainly that the Greek writing system had no way to express some of these sounds. It ended up with this (or something like it, as there is some variation from manuscript to manuscript):

ελοι ελοι λαμα σαβαχθανι
elü elü lama saḇaḥṯani

e-loo e-loo lema savakhthani

  1. In Greek, there was not a sufficient 1 to 1 relationship with Aramaic vowels. Galilean’s ə (shewa) and its open vowel a (patah) were under many circumstances differentiated solely by emphasis and were slightly colored depending upon what sounds fell nearby. In trying to approximate them, the Greek scribe chose what sounded the closest based upon Greek vocalization.
  2. The Greek alphabet has no way to indicate an “h” sound in the middle of a word, only at the beginning. So the “h” sounds  in əlahi disappeared, and there was an unintended consequence: The two letters ο (omicron) and ι (iota) when placed together formed a diphthong, similar to the nasalized eu in French. In truth, if the diphthong were broken and the two vowels spoken separately with an “h” in the middle, they are very good approximations to the original.
  3. There was also no way to express an sh sound (above š) so it was replaced with what was closest: σ (sigma, an “s” sound).
  4. There was no “q” sound, which in Aramaic is a guttural “k” in the very back of the throat. It was replaced with χ (chi, a sound like clearing your throat).
  5. And finally, the particular quality of the t was closer to their θ (theta) than to their τ (tau), so it was replaced with the former, softer sound.

Now when the Bible was translated into English, it went through yet another transliteration… but this time from the Greek. It looked (for the most part) like this:

Eloi, Eloi! Lama sabachthani?

How did we arrive at this from the Greek? Greek transliteration into English made use of the following conventions:

  1. Again, Greek vowels aren’t at all 1:1 with English vowels — they represented different sounds — but their cognates in transliteration were very well established.
    • ε and η → e, 
    • ο and ω→o,  
    • ι→i,
    • α→a,  
    • υ→y or u, 
    • etc. 
  2. The use of these transliterations actually broke up the οι diphthong in reading — so that was a step back in the right direction.
  3. The letter χ (ḥ, chi) is, like in German transliteration or Scottish, rendered as “ch,” as that digraph ch in makes a similar sound.
  4. The letter θ (theta) is transliterated as “th” as that’s the closest sound in English, although the quality of it is not nearly as breathy.

So there you have it.


3 thoughts on “My God, my God, why have you forsaken me?

  1. I won’t rule out that Jesus expected Elijah to ring in the imminent “Kingdom of God” he, as the bar nasha, preached. Perhaps, through swollen lips and tongue and with blood and fluid filling his lungs he did indeed cry out Eliya, eliya lama shevawktany “Elijah, Elijah why have you deserted me” rather than calling to God as in Psalm 22 Elohy, elohy lama shevawktany but this is a secondary position to my belief that the Aramaic cry from the cross meant exactly what it says and Jesus didn’t screw it up, Matthew did!

    I think Mark was correct in his transliteration since the Western Aramaic
    (Judean) would have a qamets qatan instead of qamets gadhol for the lamed in alaha. Easterm (Syriac) would be alef (pattah)-lamed (qamets gadhol)-heh (hiriq qatan)-yod, hence aLAhy. Western (Judean) would be alef-lamed (qamets qatan)-heh (hiriq qatan)-yod, hence aLOhy, hence Mark’s transliteration as ELWI. Judean Aramaic aLOhy, aLOhy LAma shevawqTAny? “God of me, God of me, why have forsaken you me?”

    Is there a problem with the absence of a smooth breathing for the transliterated ELWI? I don’t think so. There was no such thing in the first century and the original Markan autograph would have had an uncial ELWI.

    Polytonic minuscule did not begin to appear in the papyri until the 2nd century CE but the aspirant was disappearing in popular koine and in Egyptian Greek by the 1st century. The exact dates of these phonetic changes are probably uncertain but the aspirant appears to have disappeared in some usage by the Late Roman and Early Byzantine period. It was being retained in “learned Greek” (pre-Koine Attic) but I think we both agree that the first Gospel is not learned Greek.


    1. It may be interesting to consider how Dr. George Lamsa looks at this passage. He would actually translate it as, “My God, My God, for this I was kept.” He suggested that had the intention been to quote the Psalm 22 passage in Aramaic, Jesus would have uttered “nashatani” instead of “sabachthani” to use his phonetic spelling. Cite the book: “Idioms In The Bible Explained And A Key To The Original Gospels”, George M Lamsa. HarperCollins Publishers, NYC, (c) 1985, Pp. 102-104.

      1. Lamsa’s conjecture is simply wrong (along with a large number of his other pet theories in his “Idioms”).

        Where שבק /šbq/ in this sense is well-attested (and is also the choice made in several Targums) the verb נטש /nṭš/ isn’t clearly attested in Galilean Aramaic – if it appears at all. It appears very late in Late Jewish Literary Aramaic, far after Jesus’ time.


Leave a Reply