I know we inherit our alphabets (including its ordering) from the Romans, and if we trace it further we will end up with the Phoenicians or some other civilizations in the ancient Middle East. Do (or did) the alphabet sequence have any meaning/sense in it? For example, does it follow some rules based on the letter's phonology, etc. How were the positions of the additional letters determined, do they follow the same conventions/rules?

  • I will look up a source for you later and (hopefully) answer the question more fully, but I remember reading that (perhaps it was in Greek?) the letter values also corresponded to numerical values at one point. Of course, through borrowing and the loss/addition of certain letters this system would have become opaque -- so I'm not sure if that explains the current ordering.
    – user325
    Commented Nov 4, 2011 at 19:12
  • @Knitter That answer would be problematic for two reasons: It leads to a chicken-egg question, and even if the numerals were the source of the alphabet, the ordering of numeral words is completely arbitrary
    – Dan Milway
    Commented Nov 5, 2011 at 0:33
  • 1
    1. I believe there used to be two orders, neither of which has any point known to us. 2. @Knitter: The Greeks just used the letters of the alphabet for numbers in the order as they appeared in the alphabet. The numbers were based on the alphabet, so it explains nothing about the ABC order.
    – Cerberus
    Commented Nov 5, 2011 at 4:23
  • (On the other hand, the use of letters as numerals might have created pressure to keep the letters in a single, stable, universal order instead of rearranging them or treating them as an unordered collection.) Commented Nov 5, 2011 at 22:37

5 Answers 5


The original alphabet

We do not appear to have any ancient authority on why alphabetical order is the way it is.

Several modern scholars, such as Cyrus Gordon and Alessandro Bausani, have posited that the order is in various ways based on the calendar or the lunar zodiac. (Giovanni Garbini gives a summary of Bausani's theory in his essay The Question of the Alphabet; this is quoted in many places online.)

Writing systems scholar Peter T. Daniels, however, has stated that there is no evidence for such suggestions, and has also stated:

Scholars have been trying for centuries to explain the order of the letters of the alphabet, and no plausible answer has ever been found.

Additional letters

As for additional letters, there is no hard and fast rule that I'm aware of, but I believe the most common practice has been simply to add them to the end of the alphabet.

The Greeks did this with most of their new letters; beyond tau (the last letter of the Phoenician alphabet) are added upsilon, phi, chi, psi, and omega. When the Romans invented the letter G, they put it in the gap that had been created by dropping Z from its original place between F and H. However when they borrowed Y and Z from Greek later, these new letters went to the end of the line.

If the new letter is derived from an existing letter, it may sometimes be added to the alphabet immediately after it (thus J follows I, V and W follow U, Ñ follows N), but even in these cases this does not always happen; for example, the letters æ, ø, å, ä, and ö used in Scandinavian languages are added to the end of the alphabet.


Although the reason may be unknown, the Phoenician alphabet has enough phonetic patterns to support the idea that alphabetic order was at least partly influenced by phonetics. (That is, it would be unlikely for a phonetically-independent order to show this many phonetic patterns.) Some of these patterns are still observable in modern alphabets.

  • the nasals are grouped together (m,n)
  • (moreover, 3 of the 4 liquids are grouped together (l,m,n, though not r))
  • all 3 voiced stops are grouped together (b,g,d)
  • 3 of the 5 voiceless stops are near each other (p,_,q,_,_,t, though not k or ṭ)
  • the sequence p,_,q,_,_,t has the same place order as b,g,d (labial, velar, dental)
  • 1
    This is all wrong (see my answer, below).
    – fdb
    Commented Aug 25, 2014 at 17:28

The oldest known form of the North Semitic alphabet is the Ugaritic alphabet. The order of the letters is:

ʾa b g ḫ d h w z ḥ ṭ y k š l m ḏ n ẓ s ʿ p ṣ q r ṯ ġ t ʾi ʾu s2

As you can see, the voiced stops b g d and the liquids/nasals l m n are NOT grouped together without a break. There is no perceptible logic to this order.


it is interesting to list the letters in this way:

A b c d E f g h I j k l m n O p q r s t U v w x y z

this is slightly more ordered than we should expect from pure chance. it does not work as well if we do it in a more early-Roman way:

A b c d E f h I k l m n O p q r s t U x just mentioning this here because i believe it may be interesting. if anything, the alphabet has grown a tad more orderly over time through the addition of new letters and their respective placements.

  • 2
    What order do you see there? Explain it please!
    – curiousdannii
    Commented Jan 27, 2018 at 22:33
  • Each vowel starts a sequence of three or five consonants before the next vowel shows up. You can read them out in augmented groups of three as in A+(bcd), E+(fgh), (Ijk)-(lmn), (Opq)-(rst), (Uvw)-(xyz). If you do so (and use a monosyllable like 'we' to read out w) you get a very rhythmic rendering of the alphabet. Commented May 20, 2021 at 23:42

I find a reason for the ordering of sequence in English alphabet system.

For instance take U and V. Transformation of (letter no. 21) U-shaped curve (flat, scattered below) to (letter no. 22) V-shaped curve (steep) in many economic, social, and behavioral cycles is gradual, process demanding consistency (unity of effort) so that downward curvature polishes to a point after iteration meaning operation accomplished or victory achieved. From shape perspective, V is then improvement, perfection of U.

Chances are that most will not agree with my line of thinking. As a result, I am not posting here any more of my observation but would love to do so if there is any word of support.

  • 4
    There is a chance that people might agree with your opinions. These chances will tremendously increase if you try to build on previous research, and argue why or why not you follow their arguments, or how you can modify them.
    – robert
    Commented Nov 16, 2014 at 14:23
  • 4
    @BagraRajeev you're getting all these downvotes because you're pushing outlandish theories with no shred of evidence. If you are interested in historical linguistics, you can make a good start with amazon.com/Trasks-Historical-Linguistics-Larry-Trask/dp/…
    – prash
    Commented Nov 16, 2014 at 17:46
  • I fail to understand why a truth say 2+2=4 has to be backed by historical evidence. If speaking and listening is part and parcel of we as human beings, what's the point in finding meaning with words while correlating with how our ancestors used the same? Commented Nov 18, 2014 at 8:59
  • 1
    @BagraRajeev "2+2=4" and "downward curvature polishes to a point after iteration" are not even remotely comparable.
    – prash
    Commented Nov 18, 2014 at 23:41
  • 1
    @prash W or Double U. If U=V, then after U, X which requires two entities for multiplying is natural and branching equates Y in ZigZag manner. So, there is a reason for ordering from U to Z. Commented Nov 20, 2014 at 5:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.