Saturday, July 14, 2012

Abuse of alphabets in Jonathan Sack’s The Great Partnership

Now that I’ve gotten over my shock and horror (I, II, III), here is a straightforward summary of the errors that underlie one of the central arguments of the chief rabbi’s attempt at a grand synthesis of spirituality with science.

Sacks’ argument runs that Hebrew (being written right-to-left and without vowels) engages the right hemisphere of the brain (and so engenders or conduces a right-brain worldview), whereas Greek (being written left-to-right and with vowels) engages the left hemisphere (and so engenders or conduces a left-brain worldview). Hence, Sacks asserts, removing the Hebraic (Judeo-Christian) element of our society makes it, and us, one-sided, unbalanced, unstable: wholesome societies are comprised of whole people who use the left and right hemispheres of their brains.

What follows is simply a list of three types of errors in his writing-systems-based argument: failures of fact, of unconsidered alternatives, and of “big picture” synthesis. I leave you draw your own conclusions.

But please don’t read this as: “Oh, Sacks got his linguistics wrong. Guess I’d better skip that bit and read the rest of the book.” The questions the list of errors raises are: How difficult would it have been to avoid these errors? How likely is that The Great Partnership really understands and values science and scholarship? Is it really qualified to offer “big picture” solutions concerning society, knowledge, and meaning?

You can’t fight off a locust swarm with a fly swat. So, if I concentrate on the abuse of alphabets, it’s not because there aren’t problems elsewhere. I’m just tending to my own backyard. (Not that I’m ruling out adding in another couple of cents’ worth later…)

Errors arising from failure to check or reflect on basic facts

1. Hebrew does not write without vowels.

EVIDENCE. If so, gamal (camel), gamul (weaned), gimel (letter ג), and gomel (benefactor)—and gmala, gmali, gmalo (her/my/his camel), etc.—would all be written identically. They aren’t.

2. Presence of vowels does not mean that Greek (English) is read letter by letter, as opposed to ‘big picture’-wise reading for “vowelless” Hebrew.

EVIDENCE. The closest we come to letter-by-letter reading is when we encounter unknown words, like isobutylparaben. Otherwise, we recognise words in a single glance. This is the basis of the common psycholinguistic/neurolinguistic experimental technique of ‘masked priming’, in which subjects read words that are flashed at them so quickly that they are not aware of having seen anything.

3. Reading in Greek (English) versus Hebrew (Arabic) does not show opposite lateralization.

EVIDENCE. See the left lateralization (left inferior fusiform gyrus and left middle fusiform gyrus) of brains reading Hebrew and English here. (Besides a lit review, I followed up with researchers in Jerusalem, London, Paris, and in the joint New York University – Abu Dhabi research team.)

4. Purported generalization. Voweled scripts run left-to-right. Vowelless run right-to-left.

COMMENT. The Great Partnership asserts that the opposing voweliness and direction of Greek and Hebrew cannot be coincidence. If this statement is meant to cover just Greek and Hebrew, then it is a basic error of reasoning: with only two data points, any correlation could be a coincidence. The statement only has content if taken as a generalization, in which case it is clearly false.

EVIDENCE. Voweled scripts not running left-to-right: Ladino, Meroitic, Mongolian, Sogdian, Sorani, Thaana (Maldives), Yiddish. Vowels are largely predictable in Iatmul, Kalam, Maltese, Yimas, yet these are not written right-to-left.

Errors arising from failure to consider alternative explanations.

5. Ancestry and ambient systems, not vowellessness, account for direction.

EVIDENCE. Hebrew and other Semitic scripts inherit right-left direction from Proto-Sinaitic, which inherited it from hieratic. Voweled right-left scripts with inherited/borrowed direc- tion: Ladino (from Hebrew), Meroitic (from Egyptian), Thaana (from Arabic), Yiddish (from Hebrew). Similarly, Iatmul, etc. inherit direction from Latin (or daughter systems). Vertical inheritance or borrowing from Chinese (or daughter systems): Japanese, Korean, Manchurian, Mongolian, Old Uyghur, ’Phags Pa, Sogdian.

6. Being an Afroasiatic (Semitic) language accounts for the capacity substantially to underrepresent vowels (what TGP misnames “vowellessness”).

EVIDENCE. In all of the following cases, parent scripts are used for Afroasiatic languages, daughter scripts for non-Afroasiatic ones: Greek (Hellenic) added vowels to the Phoeni- cian script; Ladino (Romance), to Hebrew; Meroitic (isolate), to Egyptian (hierogryphs and hieratic); Sogdian (Iranian) to Syriac; Sorani (Iranian), to Arabic; Yiddish (Germanic) to Hebrew.

7. Underrepresentation of vowels in Afroasiatic languages is derivative of a grammatical quirk (not of directionality, or other causes).

EVIDENCE. Afroasiatic languages have the property that prefixes/suffixes predict (to within a small margin of error) the vowels of the root. E.g., m (the present participial prefix; cf, English ing), prefixed to a verb determines the missing vowels (e.g., MSPR = mesaper; MTYL = metayel; MGMZ = megamez, etc.; or, if t is also present, e.g., MTPRD = mitpared; MTGDL = mitgadel; MTRSN = mitrasen; etc.).

COMPARISON I. Predictable material is generally omitted in writing systems. E.g., English does not represent the change in vowels triggered by ity in rapid ∼ rapidity (rapəd ∼ rəpid-ity), nor the stress shift (rápid ∼ rapíd-ity), nor the aspiration (rapid ∼ raphid-ity).

COMPARISON II. English, like most non-Afroasiatic languages, does not have affixes that deter- mine the vowels within the verb. Consider Wh’s dltng? The verbal suffix ng can only have its vowel filled in in one way, ing. But this tells us nothing about the vowels missing from the verb root: deleting, dilating, diluting, adulating, …?

Global, cumulative error

8. It is meaningless to compare Greek and Hebrew in point of vowels and direction.

REASON I. Hebrew direction is derivative of older systems. So, you should compare Greek with Proto-Sinaitic, or the Egyptian of Amenemhat III. Comparing Greek and Hebrew is as meaningful as comparing English/Yimas with Egyptian.

REASON II. There is no reason to believe that there is any significance to Greek direction. Where there is little or no pressure to preserve or adopt a given direction and/or the system is radically redesigned, there may be experimentation with alternative directions. Such experimentation may happen in the absence of vowels (e.g., Sabean).

REASON III. Presence/absence of vowels correlates with the grammatical properties of the linguistic families to which Greek and Hebrew belong.


  1. Hmm, I am not convinced by your arguments. You are certainly ignorant about the fact that Sacks was referring to the Ancient Hebrew used in the Tanak (Old testament). For instance, that’s why that people don’t know the tetragrammaton IHVH was originally pronounced and that’s why IAWEH and IEHOVAH are two different versions for translating God’s name in the Bible.

    Perhaps he should've mentioned also that the usage of allegories, stories, parables, music and poetry does in fact emphasise a different way of thinking when you constrast the "Hebrew Prophets" with the "Greek philosophers". One can clearly see this when comparing readings of Isaiah or Ezekiel with Plato and Aristotle.

    1. Thanks for your comment, but, as stated, I have difficulty seeing that it is either true or relevant (particularly the first sentence). Which of my arguments do you think relies crucially on modern as opposed to ancient Hebrew? Can give page references (or references to interviews) in which Sacks crucially relies on the distinction between ancient and modern versions of the language?