Oh! Oh, selah!

A picture of an elderly King David playing a harpNote: This is a repost, by request, of a post originally published on June 27, 2007. (If you need to know why it disappeared in the first place, please read about the “Secret origins of the Higgaion reboot.”)

In a recent June 27, 2007 post, Rick Brannan expressed a desire to read what John Hobbins, Tyler Williams, and I might have to say about the somewhat mysterious word סלה selah, which appears some 71 times in the psalms, plus three more occurrences in Habakkuk 3. I do not pretend to any special knowledge about this word, but in keeping with Rick’s request, I will share some observations.

First, etymology, which I think is not much help. Lexicographers recognize two verbs spelled סלה. One is translated “to value” and the other “to reject.” There is also a verb סלל “to build up.” I have no idea whether any of these are related etymologically to the mystery word סלה.

Now, for the function of סלה. When asking questions like this, I prefer to work inductively, so here’s what I notice about how סלה is used in the psalms.

Sometimes, סלה seems to occur in a position where the psalm content seems to shift mood, or to break up “contrasting” content. For example (all quotations are NRSV, and the English verse numbering is used; the Hebrew verse numbering will often differ by one verse, since English versions don’t number the superscriptions, while the Hebrew text treats these as separate verses):

Ps 3:2–3 …many are saying to me,
“There is no help for you in God.”
But you, O LORD, are a shield around me …

Ps 4:2–3 How long, you people, shall my honor suffer shame?
How long will you love vain words, and seek after lies?
But know that the LORD has set apart the faithful for himself,
the LORD hears when I call to him.

See also Ps 4:4–5; 7:3–8; 32:4–5, 5–6; 39:11–12; 44:8–9; 46:3–4 (?), 7–8 (?); 52:5–6; 54:3–4; 57:6–7; 62:4–5; 89:37–38 (praise/סלה/lament).

Other times, סלה seems to occur in a position between distinct but consistent thoughts. Some of these could be seen as סלה standing between a more specific statement and a more general statement, but this pattern is not evident in all of these examples:

Ps 3:4–5 I cry aloud to the LORD,
and he answers me from his holy hill.
I lie down and sleep;
I wake again, for the LORD sustains me.
(I.e., God answered my prayer on a specific occasion / סלה / I trust God in general.)

Hab 3:3 God came from Teman,
the Holy One from Mount Paran.
His glory covered the heavens,
and the earth was full of his praise.
(I.e., God was made manifest in a specific location / סלה / God was made manifest in general.)

Hab 3:9 You brandished your naked bow,
sated were the arrows at your command.
You split the earth with rivers.

Hab 3:13–14 You crushed the head of the wicked house,
laying it bare from foundation to roof.
You pierced with his own arrows the head of his warriors …

See also Ps 9:16–17 (where you will also find another interesting term); 20:3–4; 21:2–3; 39:5–6; 49:13–14; 52:3–4; 55:7–8; 57:3; 60:4–5; 61:4–5 (desire/סלה/rationale?); 62:8–9; 67:4–5; 68:7 (chronological setting/סלה/events), 19–20, 32–33; 75:3–4; 76:9–10 (?); 77:9–10, 15–16; 81:7–8; 82:2–3; 83:8–9 (complaint/סלה/plea); 84:4–5, 8–9; 85:2–3; 87:6–7; 88:7–8, 10–11; 89:45–46 (complaint/סלה/plea), 48–49 (complaint/סלה/plea); 140:3–4 (complaint/סלה/plea); 8–9; 6–7.
At still other times, סלה is the last word in the psalm (see Ps 3:8; 9:20; 24:10; 46:11).

One could pretty easily mount an argument that סלה separates stanzas or liturgical “movements” in Psalm 24. However, “exporting” that understanding to other psalms doesn’t work so well, and requires a great deal of speculation. The same explanation might work for Ps 32:7–8, where סלה stands between address to God and address to the human audience, and Ps 47:4–5; 48:8–9; 49:15–16 (change of addressee?); 50:6–7 (change of speaker); 66:4–5 (change of addressee), 7–8 (change of addressee), 15–16 (change of addressee); 67:1–2 (change of addressee); 76:3–9 (change of addressee); 77:3–4 (change of addressee); 87:3–4 (change of speaker); 89:4–5 (change of speaker). In Ps 59:5–6 and 59:13–14, סלה stands at the end of a verse, positioned just before the Psalm’s refrain.

The use in Ps 55:19, where סלה seems to interrupt the sentence, is utterly opaque to me.

So, I perceive several different positions in which סלה appears:

  • Topic A / סלה / topic B (contrastive)
  • Topic A / סלה / topic A’ (consistent)
  • Speaker A / סלה / speaker B (change of speaker)
  • Addressee A / סלה / addressee B (change of addressee)
  • סלה before refrain
  • סלה at end of psalm
  • סלה at change of liturgical action

I cannot really see any particular “common thread” that unites all these uses. In short, I’m baffled. I know of no better explanation than those that place סלה in the company of words like הגיון ,שגיון ,משכיל, etc.; that is, mysterious words that probably have some kind of liturgical or musical significance that is now all but lost to us.

Sorry, Rick. It’s not very satisfying. But it’s the best I can do just working with the Hebrew text. I’m unaware of any use of סלה in any inscription or whether it appears in the “sectarian” poetry from Qumran, and I’m not in a position (geographically) to check at the moment.


7 thoughts on “Oh! Oh, selah!

  1. Susan Burns

    There is another word that sounds like selah and means “rock” but is spelled differently. However, if both words were in existence before invention of writing, how would they be differentiated?

    1. Chris Heard Post author

      Most simply, speakers of Biblical Hebrew would differentiate סלה and סלע (“rock”) the same way anybody else differentiates between homophones: by context. When a skilled English speaker hears the sound /rĕd/, he or she has no trouble differentiating between the color red and the past-tense verb read. Speakers of English easily parse sentences like “Have you ever read Mao’s Little Red Book?” without confusing the two senses. Speakers of all languages, including Biblical Hebrew, handle homophones in exactly the same fashion.

      But more precisely, סלה and סלע were probably not actually homophones in ancient oral Hebrew. In סלה, the letter ה is a “vowel letter,” an inaudible indicator of the preceding vowel sound in a language without written vowels. Despite the practice in modern Western classrooms, though, the letter ע is not really a “silent” consonant, but rather an audible pharyngeal stop, like a gulping sound at the back of the throat—almost a /g/ wannabe. Clear evidence for this comes from the ancient Greek translations of Hebrew biblical texts, in which proper nouns starting with ע start with Γ in Greek: עַזָּה becomes Γάζα, “Gaza,” for example. Furthermore, in the Masoretic vocalization at least, the two words in question have different second-syllable vowels. So an ancient Israelite would have no trouble distinguishing between our mystery word סלה, pronounced like /sĕ-lâ/, and the word סלע, “rock,” pronounced more like /sĕ-lărgh/.

  2. Susan Burns

    With my limited understanding of afro-asiatic languages, it appears to me that words coming from the same root are somehow related even if we are unable to ascertain what that relationship could have been. There is a word “selad” that uses the same samekh-lamed root (sorry I don’t have Hebrew font) and means “exalt”. This might be a better example for me to use than selah (rock). Could this word give us a clue as to the meaning?

    1. Chris Heard Post author

      No, this line of reasoning doesn’t yield profitable results.

      First, Biblical Hebrew roots (and many Semitic roots more generally) are usually three letters; there is no verbal root סל in Biblical Hebrew.

      Second, a verb spelled סלד appears only once in the entire Tanak, in Job 6:10. Its meaning is uncertain, though usually regarded as meaning “jump.” It’s dangerous to base any conclusions on a word of uncertain meaning that appears only once in our corpus. (There is also a proper noun סֶלֶד, but that’s irrelevant here.)

      Third, there is a root סלל that can signify exalting or praising, but it can also signify piling stuff up (like sheaves of wheat, or stones to serve as a siege-ramp). This the root that gives us סֻלָּם, Jacob’s “ladder/stairway to heaven” in Genesis 28. Some investigators have speculated that סלה might derive from סלל and might mean “raise the pitch” (like the key changes so common in many modern praise songs the last time through the chorus), but this requires emending the Masoretic vocalization from סֶלָה to סֹלּוּ—a fairly significant change—and is speculative anyway. Deriving סֶלָה from a root סלל is not at all straightforward. The simple occurrence of a shared סל tells us nothing about etymology. The transformations of roots into nouns follow predictable patterns, and none of those patterns lead from סלל to סֶלָה, although one does lead to סֹלְלָה, “siege-ramp.”

      Instead of grasping at straws, it’s probably better to admit the limitations of our knowledge, given the data available to us at this time.

    1. Chris Heard Post author

      “Must be” is surely an overstatement, in light of the facts that (a) deriving סלה from סלל is speculative, (b) deriving סלה from סלל requires emending every occurrence of סלה in the MT, and (c) there is no evidence other than this speculative derivation to suggest that סלה refers to a motion or posture. If סלה is a verb derived from סלל, it is definitely not an imperative form, but then סלה doesn’t appear to be a verb form at all (certainly not using the Masoretic vocalization). סלה “must have been” meaningful to those who compiled the psalter, but its meaning is opaque to us given our current level of knowledge—which will only improve with new data, not with speculations.


Leave a Reply