Why are the names of some letters hard to understand on the telephone?

Question

It is common experience that some letters are hard to understand when spelling a word in a telephone conversation.

What exactly is the reason for this and is there any research as to which letters this concerns?

Apparently this phenomenon was discovered early on when radio telecommunications became common.

It was the reason why spelling alphabets were invented. The British Army began using a spelling alphabet in 1898:

British Army signallers began using a partial spelling alphabet in the late 19th century. Recorded in the 1898 "Signalling Instruction" issued by the War Office and followed by the 1904 Signalling Regulations this system differentiated only the letters most frequently misunderstood: Ack (originally "Ak") Beer (or Bar) C D E F G H I J K L eMma N O Pip Q R eSses Toc U Vic W X Y Z.

(source)

Here, the following English letters are listed as "most frequently misunderstood":

• A (?)
• B (probably in contrast to P)
• M (probably in contrast to N?)
• P (probably in contrast to B, listed above)
• S (probably in contrast to F? - but this is just a guess)
• T (probably in contrast to D?)
• V (probably to distinguish from W?)

The following source gives an example of six letters which are hard to distinguish because they are articulated in a more or less similar manner, but doesn't give a clear explanation for the reason behind the phenomenon:

letters with a similar manner of articulation can be virtually indistinguishable, for example, nasal consonants like “m” and “n”; fricative consonants like “f” and “s”; stop consonants like “t” or “d”. Vowels, on the other hand, carry through relatively clearly

(source)

It seems, however, that the phenomenon is mainly concerned with consonants.

According to yet another source, the pass band of telephones is limited to frequencies ranging from 300 to 4000 Hz:

The pass band of telephones is typically about 300 to 4000 Hz, so the fundamental is usually absent or much attenuated. The loss of information carried by frequencies above 4000 Hz (e.g. the confusion of ‘f’ and ‘s’ when spelling a name) is noticed in telephone conversation, but the loss of low frequencies is much less important.

(source)

This might explain why some sounds are hard to understand, because they are virtually cut off.

This and this audiogram show that "f" and "s" actually are above 4000 Hz -- on the other hand, it shows that d and t just like b and p are well between 375 and 4000 Hz, so this still doesn't answer the question why those consonants are hard to distinguish.

Answer 1

The language is always excessive, that is, it helps to understand the speech in several ways at the same time ­- there are other words that create the context, there is the grammar an the syntax of the sentence that help us understand the exact meaning of the words we hear and not to mix them up with their homonyms, also many languages, like English, have lots of dissimilative spellings that, again, help us to see which homonym is used, e. g. whole or hole, write or right or rite, etc. And when people speak, we can read the lips of the speaker which also greatly helps in understanding, since many consonants sound really very similar.

Now imagine, when the radio appeared, you could now hear a person without seeing him or her, and when the words were spelled letter-after-letter, almost all that excessive features were gone, not to say about the quality of the sound on the radio (and even very often on the modern phones), so that 1898 spelling alphabet was created so as to add more excessiness to the names of the letters.

Also, there's such thing as McGurk effect that shows how little is needed to make us hear a sound absolutely different from the one that was pronounced. Here's a great BBC video (3 minutes) that demonstrates the McGurk effect, watch it and you will see what I mean: https://www.youtube.com/watch?v=G-lN8vWm3m0