two-letter roots in early Semitic language(s)

Discussion in 'Etymology, History of languages, and Linguistics (EHL)' started by Josh_, Feb 2, 2008.

  1. Josh_ Senior Member

    the phrontistery
    U.S., English
    This post is long, but basically, I just wanted to get some your opinions on the (possible) existence of pre-Semitic, or proto-Semitic, two-letter roots, that ended up becoming three letter roots in the Semitic languages. Do you think it is possible that two-letter roots existed? If not, how could we explain why all these words have related meanings? What are other possible reasons that these roots with related meaning all start with the same two letters?

    I'm sure this topic has been discussed in depth by linguists, but only recently have I become somewhat interested in the roots, so to speak, of Semitic languages. If you know of any articles, books, etc. on the subject I would be interested in reading them.

    Anyway, the following is just some evidence that I have found that lends credence to the idea of two-letter roots.

    Every so often I will be perusing an Arabic dictionary and note how it seems strange that many roots with related meanings are so close together in the dictionary. I didn't give it much thought until the other day when I was reading an article about Semitic languages that suggested that there is evidence of biconsonantal (two letter, aka biliteral, ثنائي الحروف) roots in pre-Semitic that were later made triconsonantal (three letter, aka triliteral, ثلاثي الحروف ) roots. The example that the author gave was the Hebrew roots p-r-d 'divide', p-r-m 'open, seam' p-r-s 'break up, divide up', suggesting an old root p-r. The p-r here is the Hebrew פ (peh) and ר (resh) which are the cognates of the Arabic ف (faa) and ر (raa). (Even looking at the letters we can see a resemblance between the Hebrew and the Arabic.) This is interesting, because looking at entries in an Arabic dictionary that start with ف and ر is the first time I thought it strange that so many words with related meanings were so close together in the dictionary. I knew there must be some relation between them, but it never occurred to me (until reading that article) to consider the root letters. Here are the roots:

    ف-ر-ج (f-r-j) -- ideas related to opening, parting, separating.

    ف-ر-د (f-r-d) --
    ideas related to being singular, alone, separated or segregated from.

    ف-ر-ز (f-r-z) --
    ideas related to separation, isolation, detachment, differentiation, distinguishment.

    ف-ر-ط (faraTa) -- in addition to the meaning of excess this root has meanings related to separating things (into their component parts), stripping off, etc.

    ف-ر-ع -- (f-r-3) -- ideas relating to branch ing(out) or putting forth branches, in other words related parts that become separated and/or detached from one another.

    ف-ر-ق (f-r-q) --
    ideas related to separation, parting, division, differentiation, discriminating, distinguishment.

    ف-ر-ك (f-r-k) -- it means to rub, but has the additional (perhaps original) ideas of rubbing (something) until it starts to fall apart (i.e. separates).

    ف-ر-م (f-r-m) -- ideas related to cutting or mincing (meat) into small pieces (in other words detaching and separating).

    ف-ر-ي (f-r-y) -- ideas related to cutting and mincing.

    There are a few more in which there appears to be a connection with this general idea of separation (for lack of a better descriptor), but the connection is loose and debatable:

    فرس (farasa) -- to kill and tear apart, prey (of a predatory animal). Here we see the idea of tearing apart (in other words detaching and separating out).

    فرش (farasha) -- to spread out. When you spread something out you separate its parts from each other (unfolding a sheet, for example)

    فرش -- farrasha -- to brush hair, in other words to separate the individual strands of hair from each other.

    فرشح -- farshaHa -- to straddle, in other words to separate one's legs.

    As already noted there are Hebrew roots using these same root letters that have meanings similar to the Arabic meanings:

    פ-ר-ד (p-r-d) -- ideas related to separation, division, parting.

    פ-ר-ט (p-r-t) -- ideas related to separation, breaking bills into smaller monetary units.

    פ-ר-ץ (p-r-ts) -- ideas relating to breaking open or bursting into.

    All these related roots, in both Arabic and Hebrew use the same first two Semitic root letters. This does suggest the existence of an ancient biliteral root f-r/p-r.

    So, I wanted to get your opinions on the matter. Why do all these words with the same first two root letters have related meanings? Do you agree with the suggestion of a pre-Semitic two letter root.

    As an aside, I wanted to briefly discuss the connection between some triliteral roots and quadriliteral roots as it relates to the subject at hand. Many times, a letter is added to a triliteral root and it becomes a quadriliteral one with a related idea.

    One of the most obvious is the that of طمن Tammana and طمأن Tam'ana which both have the meaning of reassuring someone. The only difference is the addition of ـأ in the quadriliteral verb.

    Some other examples are:

    غرد gharida -- to warble (of a bird), i.e. sing with trills.
    زغرد zaghrada -- to utter the zaghruuTa, a drawn out trilling sound (of women in times of joy).

    عوّذ l3awwadha -- to pronounce a charm or incantantion over someone.
    شعوذ sha3wadha -- to practice magic.

    So the idea of a root letter being added to a(n already established) root is not unheard of.

    On the subject of quadriliteral roots W. Wright, in his Arabic Grammar, discusses a little about how quadriliteral verbs are formed. He lists four ways, but for brevity I will only include the one that interests me:

    A fourth letter, generally a liquid or sibilant, is prefixed, or affixed, or inserted in the middle of, a triliteral verbal form. He includes as examples شمخر (shamkhara) to be proud (related to شمخ shamakha to be high), شمعل (sham3ala) to be scattered = شمع (shama3a); جمهر (jamhara) to collect, which he says compare to جم (jamma) and جمع (jama3a).

    The last example I listed is also an interesting case as, considering the roots with related meaning, it might possibly that a two-letter root (ج-م j-m) gave rise to some different verbs with related meanings:

    جمّ jamma (to gather).
    جمع jama3a (to gather or collect).
    جمل jamala (to sum up).
    جمهر jamhara (to gather, collect, assemble).

    There is one more I can think of offhand:

    خرط kharaTa -- to pull, strip, cut into small pieces.
    خرق kharaqa -- to tear (apart), pierce, bore (a hole)
    خرم kharama -- to pierce, make holes in.

    I'm sure there are others, but I can't think of them right now.
  2. cute angel Senior Member

    the universe
    I didn't understand you well but these words exist and even two letters such as

    مح=he yallow of the egg
    أح=the white of the egg
    بن=the origin of coffee
    جن=invisible creatures like angels
  3. Mahaodeh Senior Member

    Arabic and English
    Cute Angel, I'm afraid the examples are of three letter words, each one of them has a shadda on the second letter.

    مح=he yallow of the egg - مَحّ, the root is م ح ح
    أح=the white of the egg - أحّ, the root is أ ح ح (this is the first time I hear it means the white of the egg! Maby it's local to Algeria?)
    صح=right - صَحّ, the root is ص ح ح
    سن=tooth - سِنّ, the root is س ن ن
    بن=the origin of coffee - بُنّ, root ب ن ن (I'd call it "coffee beans or coffee plant)
    كن=be - كُنّ, root ك ا ن (notice the past and present tenses كان يكون)
    جن=invisible creatures like angels - جِنّ, the root is ج ن ن

    However, there are some words of only 2 letters; but they are limited and most are actually حروف such as لا - بَخْ - في - بَلْ - ما - إنْ - لَمْ so I don't really know if these count.
  4. Josh_ Senior Member

    the phrontistery
    U.S., English
    Yes, I mean verbs and words derived from verbs. Many articles (حروف ) are only two letters.

    Actually there are two words -- أح (l2aHHa) which means to cough and آح (l2aaHun) with a مدة, meaning egg white, which appears to come from the root ء-و-ح , but, I suppose, because of the rules of إبدال the و is changed into another alif for ease of pronunciation - يعني، يتم إبدال الواو ألفًا -- in the same manner that the word 'be' from the root ك-و-ن, is rendered كان in the past tense. If آح is not the normal word for egg white, what is, in MSA?

    كن is only two letters, but is from the root ك-و-ن . And since it is in صيغة الأمر , which is مجزوم , and the root is معتل العين , the و drops out.
  5. Mahaodeh Senior Member

    Arabic and English
    Regarding egg white, I've always read, heard and used it as بياض البيض. I'm not claiming that it's incorrect, I just never heard it. I made a quick search in a few dictionaries and didn't find that meaning either. The only word I found was أحّ with a shadda.
  6. Qittat Ulthar Senior Member

    London, England
    Dutch (Netherlands)
    Josh_, I am delighted that you bring this up, because it happens to be the subject of my master thesis. :)

    I think the most well-known theories of the subject have been formed by Georges Bohas of the University of Paris. He states that Arabic roots are derived from what he calls "etymons", which are a combination of two letters, with an added third letter. The letters of the etymon can be in either sequence, and the added letter can precede them, follow them or be in the middle.

    He even takes this theory a step further to say that behind these etymons lie "matrixes", consisting of two groups of phonemes, like labials and dorsals, which share a common meaning.

    If you read French, his books are fascinating, if you do not, plenty of articles have been written about his theories and some of his own articles have been written in English, for example in the Encyclopedia of Arabic Language and Linguistics, Matrix and Etymon model. Also interesting in that work is the article "Biradicalism" by Zaborski.

    If you use Google Scholar on "matrix etymon arabic" you will find a lot about these theories. They are not new, people have been finding the same things as you and wondering the same for centuries. The three-radical root notion however always seemed to prevail. Is that because it has more truth in it, or just because it is known and convenient?

    I am trying to find out...

    Edit: I would be delighted to hear any of your opinions and thougts on this, especially from the Arabs themselves, who have been trained in language a bit: how do you regard these theories?
  7. Josh_ Senior Member

    the phrontistery
    U.S., English
    I searched in a few dictionaries also and didn't find it except in the Hans Wehr in which it appeared between أوج (highest point) and أواخي (classical plural of آخية .

    Thank you, Qittat Ulthar, for the article and book references. I will check them out.

    It's interesting that you bring up the idea that the letters can appear in any sequence because sometimes I'll notice strange things like that as well. Take, for example, the root ح-م-د (H-m-d) with meanings of praise and thanking (mainly used with God). There is another root م-ج-د (m-j-d) which has meanings related to being glorious, exalted, praised. In this case we see two of the same root letters -- 'm' and 'd'. There is also another root م-د-ح (m-d-H) which has the same root letters H-m-d but in different positions. This root has meanings related to praise and commendation, but it is used quite differently from H-m-d. There is a discussion here concerning H-m-d and m-d-H and their differences.
  8. Mahaodeh Senior Member

    Arabic and English
    I'm sorry to bring up an old subject, but I was surfing through the internet and I found an interesting related page here. It also argues that Arabic (as well as other semitic) roots are originally two-letter.

    ps. it starts by talking about Arabic roots, so you have to scroll down until you reach Proto-Roots in red.
  9. Ander Senior Member

    I had that Internet page in my bookmarks (for the grammar, not the for two-letter roots).

    Arabic is rather ancient as a Semitic language, but I think Akkadian is still older.

    So it is about Akkadian that the research should be made.
  10. Josh_ Senior Member

    the phrontistery
    U.S., English
    Really, it is about all Semitic languages that research should be made.

    Thanks for sharing that article, Maha. I enjoyed it a lot.
    Last edited: Nov 28, 2008
  11. psxws

    psxws Senior Member

    Spanish-Venezuela, English-United States
    I believe I read somewhere that فم is two-radicaled?
  12. djamal 2008 Senior Member


    صحيح ما عاد فعل كان في صيغة الأمر و هو كن و ليس كنّ لتفادي إلتقاء سكنيين لذا يحدف حرف العلة ( حركته السكون) في هذه الحالة يكون الواو

    و الممكن القول "لم يك كما ورد في القرآن

    في حالة الإعراب قد يكون في بعض الأحيان الفعل يحتوي على حرف واحد : مثلا ر و ق و عادةً تلفظ ره و قه؛
  13. Defence New Member

    Ontario, Canada
    English- Canada
    I know that roots occur with great frequency in Hebrew, but I'm still new to Arabic. ברכ is a common root for a few words relating to prayers, I believe. (I haven't had a Hebrew lesson in years)
  14. Mahaodeh Senior Member

    Arabic and English
    I'm not sure what you mean by "occur with great frequency"; but if you mean "they are very recognisable and you can practically return every word to it's 3-letter root", then it's the same case in Arabic.

    However, here we are talking about the possibility of having another layer of roots, which is two-letter roots; possibily in Proto-Arabic or even in Proto-Semitic.
  15. astlanda Senior Member

    Estonian maamurre
    It's a bit off topic, but I've read, that the words of the same origin in North Caucasian languages may consist of similar phonemes in different order, because the "proto-language", which they have been developed from had so many phonemes, that it did not matter much, if you pronounce them in wrong order. I.E. It would be very unlikely to have all those phonemes in another root.

    Though, I could not find anything about it in internet.
    I think, you can read about it from:
    Diakonov Igor M., Starostin S.A. Hurro-Urartian as an Eastern Caucasian Languages. Münchener Studien zur Sprachwissenschaft, R. Kitzinger, München, 1986

    I suppose, if his postulate is true about Arabic it may be related to the onomatopoetic origin of these stems.
    Last edited: Dec 28, 2008
  16. Abu Rashid

    Abu Rashid Senior Member

    Melbourne, Australia
    Australian English
    This is an interesting topic that I've read about before. If I remember correctly, the Semitic languages of Africa (Ethiopian perhaps??) still have some two-letter verbs, I don't think Arabic does though. However, I've heard that some of the 'tools' of Arabic language that do have two letters were perhaps originally verbs/roots whose function in language changed at a much earlier period of development of the language, and that's why they still have only two letters.

    Although there may be older written records of Akkadian than of Arabic, that doesn't mean the language is older. I've seen it stated by Semiticists that Arabic is the Semitic language that most closely resembles the theorised 'proto-semitic' language. Also the theories that Akkadians and other Middle Eastern peoples originally migrated out of the Arabian Peninsula in various waves tend to support the idea that Arabic is probably much older and perhaps even the 'parent' language of most other Semitic languages. Others just developed writing systems first, that's all.
  17. Mahaodeh Senior Member

    Arabic and English
    Now that's interesting; I was under the impression that Arabic is one of the newest of Semitic languages.
  18. Abu Rashid

    Abu Rashid Senior Member

    Melbourne, Australia
    Australian English
    The dating of languages is generally based on written records alone, as that is the only real artifact that confirms the existence of a language in a given time period.

    Records of written Arabic do not go back very far, that's why it's often been considered a younger member of the Semitic family.
  19. epchois_nai_nan New Member

    Australia - English
    Hello, this is my first post - sorry to use it resurrecting a two year old thread, but I noticed something really interesting.

    I deal with the Coptic language on a weekly basis, which is essentially the native Egyptian language which was written in hieroglyphics a long time ago and made the switch to using the Greek alphabet in about the first century BC. It ceased being spoken widely about 400 years ago, but we still use it in Church services, and the internet has allowed for something of a revival to take place.

    Anyhow, the thing is, I know that Egyptian, while not a Semitic language, is an Afro-Asiatic language, which means that Coptic, Hebrew and Arabic all share a common ancestor language in the distant past. Looking at all those words in the first post to do with spreading and dividing, I decided to check my Coptic to see if any of these roots are old enough to have developed into Egyptian as well, and this is what I found! There is debate as to whether the Greek letter Phi is used to represent a 'p' or an 'f' sound in Coptic - I've represented it as 'p' because that's the most likely answer. Either way, it still fits with the root p-r:

    'poRsh': a spread place, a couch
    'poRshi': a table spread
    'pResh': a thing that has been spread
    'poRdj': to spread
    'ma-pResh': a sleeping place (a bed spread)
    'perk': to pluck out, destroy
    'porper': to be open
    'piri': to put forth
    'pekh': to break, rupture
    'pesh': to divide

    These words are all pure Egyptian - you would find them on the walls of Phaoronic tombs. I don't think it's a coincidence that they contain the sounds 'p-r' and that they all have meanings related to spreading and dividing. I think that might be pretty good evidence for double roots, stemming back all the way to proto-Afro-Asiatic itself, far older than proto-Arabic or proto-Semitic.

    Hope you find that interesting :D
  20. quee1763 Member

    Doha, Qatar
    If I remember my Ancient Egyptian correctly, "pr" is a word for "house" or "palace" among of other things.
  21. Ustaath Senior Member

    Arabian Peninsula
    Arabic - levantine
    Moderator's Note:

    This discussion about the biliteral root ك ن potentially being shared by both the roots م ك ن and ك و ن was split from this discussion.

    I meant to say proto-bilateral root :)
    Last edited by a moderator: Jul 10, 2011
  22. lukebeadgcf

    lukebeadgcf Senior Member

    Cambridge, MA
    American English
    Are you saying that you think both these triliteral roots stem from the same ancient biliteral root? (I'm sorry, can someone move this topic to a new thread?).
  23. berndf Moderator

    German (Germany)
    Moderator note: Thread moved to etymology and history of language forum.
  24. Ustaath Senior Member

    Arabian Peninsula
    Arabic - levantine
    yes Luke, that is correct - it's been sometime since I said that -or since I've participated in the forum :)

    it seems to me that the 'meem ' - fossilizes the 'k-n' to a static state of existence.
  25. utopia Senior Member

    Israel, Hebrew
    Moderator note: Discussion of the same topic split from quoted thread.

    I read somewhere that לחם both in Hebrew and Arabic have the meaning of something that contains carbohydrates - פחמימה as in פחם.

    All these words contain ח and מ.

    all connected to heat in some way.
    Last edited by a moderator: Jul 7, 2017
  26. berndf Moderator

    German (Germany)
    That is if you believe in the theory of pre-proto-Semitic two consonant roots. The reconstructed meaning of the hypothetical two consonant root ḥm is being hot. But according to this theory the added third consonant in proto-Semitic, which differentiates the more basic two letter themes further, usually appears as second or third radical and seldom as first radical. But as there is no identifiable thematic group lḥ or lm, ḥm would the most likely candidate.

    This theory is rather speculative and far from being generally accepted.
  27. aavichai Senior Member

    Hebrew - Israel
    I also believe in that theory
    and actually i'm writing about it right now

    and there are a lot of root that i don't think someone can argue about the connection of the two letter

    this theory is not "accepted" by scholars is because they "don't believe they can find that"

    here is what Blau writes
    דוגמה קיצונית מסוג זה משתקפת בניסיונותיהם של מלומדים שונים לגזור את השורשים השלשיים משורשים בעלי שני עיצורים, בהציעם קשרים סמנטיים בין משמעויות רחוקות לכאורה. כשמעיינים במחקרים מעין אלו, קשה להשתחרר מן הרושם שקשרי הקשרים המוצעים בהם, גם אם אינם נמנעים עקרונית, סיכוייהם לשקף מציאות כה מצומצמים עד שהזהירות מחייבת שלא להציעם כלל.

    and he may be right,
    but like i said, there are words that i don't think someone can argue about them

    i think the words that are accepted are just the words with close definitions as
    גזז גזר גזם...

    but i think also the words with far definition that one can find a basic definition as the mother of them are right as well

    i saw people write about it and try to bring the basic definition
    and a lot of them are not "proffesional" in my opinion as people just throwing meanings from their mind
    but i believe (and work about it) that a basic definition can be find with the right mind.
  28. Drink

    Drink Senior Member

    New England
    English - New England, Russian - Moscow
    I think basically, no one denies that there is a some kind of connection between certain roots. But nevertheless, the exact details of how or why these roots are connected are not clear. So the connections are accepted, but there is no accepted theory to explain them.
  29. berndf Moderator

    German (Germany)
    As far as I can see, that sums up the current state of the academic discussion quite nicely.

Share This Page