Reform Review: Benjamin Franklin's Phonetic Alphabet

Thank you to jan GameCuber 
for requesting this episode! 
Welcome to Reform Review, the show that gets 
facts wrong about your favourite language reforms!  
I’m the Casual Conlanger, and in this 
episode, we’ll be looking at  
Benjamin Franklin’s Phonetic Alphabet, or “bfpa” for 
short. That doesn’t sound great… but… 
I’ll figure something out by the end of the video.
Born in 1706, Benjamin Franklin was a Founding  
Father of the United States, a polymath, 
and most importantly for us, a spelling  
reformer. He proposed his phonetic alphabet in 
the mid-eighteenth century as a global spelling  
reform for the English language, though some 
reform purists might argue that it isn’t one  
in the truest sense – we’ll see why in a moment. 
It predates (yes I know I pronounced this wrong)
the IPA by over a century, and reveals  
several differences between eighteenth-century 
and modern English pronunciation. 
First, graphemics. Franklin’s Alphabet removes 
six letters ⟨c j q w x y⟩, and
adds six custom symbols, which goes beyond a small 
minority of reformers’ definition of a “spelling”  
reform, but personally, I think it counts.
Showing these added letters is a bit difficult  
in practice, as they’re not supported by Unicode. 
Thankfully though, there are a few documents on  
the Wikipedia page that we can use for reference. I’ll 
have to use a few provisional ones that somewhat  
resemble them, namely Latin alpha ⟨ɑ⟩, heng 
with hook ⟨ɧ⟩, eta ⟨η⟩, and Armenian ini ⟨ի⟩. 
Let’s IPA the FPA and have a look 
at its phonology and orthography. 

The oa-ligature stands for /ɔː/, upturned
‘h’ called “uh” represents /ʌ/,
“ing” for /ŋ/, “edh” for /ð/, “eth” for
/θ/, and “ish” for /ʃ/. The last
three all resemble the letter ⟨h⟩, likely
to symboli[s|z]e the digraphs they replaced.
Spelling is mostly phonemic, and
we’ll find out why in a minute.
So, the consonants. The first thing to note is
the lack of the wine-whine merger in the reform,
meaning words like “which” and “whine” were
distinguished from “witch” and “wine” by the
pronunciation of ⟨wh⟩, which would have roughly
been [ʍ] for Benjamin Franklin. This sort of
distinction is somewhat rare nowadays, but was
relatively widespread in the eighteenth-century,
so its inclusion is unsurprising.
Just to clear things up from previous videos,
I won’t be analysing */ç/ as a phoneme anymore…
as some of you in the comments pointed out: it’s
wrong. However, I will maintain /x/ as a marginal
phoneme, which was left out in this reform.
The reform has no distinct graphemes for /j/
and /w/, opting to use their vowel equivalents
instead, sort of like Classical Latin. This
is pretty unintuitive: take how Julius (or
/julius/) was spelt during Caesar’s lifetime…
it’s quite confusing if you’re not used to it.
Let’s turn our attention to this bizarre-looking
cell, and try to understand what’s going on here.
The first letter is called “long s”, which you may
have seen in very old texts and signs in England,
and is originally the first half of Eszett in
German. It went everywhere except word-finally
where a small ⟨s⟩ was used, eventually
abandoned in printing and type foundries by
the nineteenth-century, and in handwriting half a century later,
so we’ll just use small ⟨s⟩ for clarity’s sake.
Remember how I said that spelling in
the BFPA is mostly phonemic? Well,
⟨s⟩ was used for /s/ and sometimes word-final
/z#/, and that “sometimes” gives us the
opportunity to look at some morphology!
A morpheme is the smallest abstracted
unit of meaning in a given language,
analogous to phonemes, which are the
smallest abstracted unit of sound in a language.
For example, the word “unbreakable” contains
three morphemes: ‘un-’, ‘break’, and ‘-able’.
Since it can stand independently in a sentence,
‘break’ is called a free morpheme, whereas ‘un-’
and ‘-able’, called bound morphemes, are dependent
on free morphemes to keep their meaning.
Morphemes can be written in the language’s
orthography, the IPA or even glossing
abbreviations for clarity, and they often have
several different forms called allomorphs (like
the distinction between phonemes and allophones).
For instance, the English plural suffix
morpheme ‘-s’, usually written as //-z//,
has multiple allomorphs. It becomes /ɪz/ after
sibilants, devoices to /s/ after voiceless sounds,
and remains /z/ elsewhere. An example of each
of these is “kisses”, “cats”, and “dogs”.
In this letter written in Franklin’s Alphabet –
oh, a bit bright, there we go – the cases where
it’s pronounced /s/ or /z/ are written as ⟨s⟩, but
/ɪz/ is spelt ⟨is⟩ or ⟨iz⟩ a bit arbitrarily, and,
while we’re at it, some words such as ‘to’
and ‘be’ seem to have multiple spellings. If
this doesn’t scream “Confusing and difficult
reform even for eloquent English speakers”,
then I honestly don’t know what does.
While we’re still on the topic of morphology,
also notice the very Shakespearean ⟨’d⟩ past tense morpheme (technically a past participle morpheme here), which represents /t/ and /d/ after
voiceless and voiced phonemes respectively, and occasionally
/ɪd/. I actually quite like this feature, though I’m
not getting nightmarish flashbacks to secondary
school English classes, so that’s probably why.
You might be wondering why I’m rambling on
about morphology so much. Well, a worthwhile
advantage of incorporating some morphological
spelling rules into a largely phonetic reform
is that it can heavily simplify inflection and derivation.
Okay, enough with morphology, back to orthography.
Ish is used in all of these graphemes, which is
a neat feature, but makes me question the choice
of ⟨i⟩ for yod, since /ʃ/ could have been written
⟨sի⟩ with yod as ⟨ի⟩, although its name would
probably need changing. It feels like a missed
opportunity to improve internal consistency, and
maybe make the spelling a bit more intuitive.
Alright, consonants done, let’s move onto vowels
now! This table is messy, and for good reason: the
few examples of Franklin’s Alphabet are all rather
inconsistent spelling wise, and different sources
have made varying analyses of the orthography,
so… I bundled them all up in one and… well,
it’s clear that Franklin didn’t spend much time
on the vowels when creating the reform. Above all,
it’s vague: with one or more phoneme assigned
to each grapheme, these choices complicate
the vowels just as much as they simplify them.
It doesn’t get much better with the diphthongs.
Given how old the reform is, I can cut some
of Ben’s decisions some slack: for instance,
the choice of ⟨ee⟩ for /ei̯/, which was probably
pronounced more like [eː~ɛː] in the 18th century,
but that’s just about the only positive thing I
can say about the diphthongs. They’re appalling.
Like, how is ⟨ɑɥi⟩ a good representation for
/oi̯/? I just don’t know. Why is the /a/ in /ai̯/
represented differently from the /a/ in /au̯/?
Now, admittedly, the vowel choices could be
because of his accent, but that would make
them diacentric! Every way you look at it,
for what’s supposed to be an
intuitive global spelling reform,
the vowel graphemes are the total opposite.
As always, let’s look at the reform in
practice. Pause the video and have a go at it,
and let me know how you did in the comments!
“Kensiŋtɥn, 26 Septembɥr, 1768.
Diir Sɥr, ɥi hav transkrɥib’d iur alfabet,
&c., huitի ɥi ηink mɥit bi ɑv sɥrvis tu ɧos, ho
uiի to akuɥir an akiuret pronɥnsieիɥn, if ɧat kuld
bi fiks’d ; bɥt ɥi si meni inkɑnviiniensis, az uel
az difikɥltis, ɧat uuld atend ɧi briŋiŋ iur letɥrs
and ɑrηɑgrafi intu kɑmɥn ius. ɑɑl ɑur etimɑlodիis
uuld be lɑst, kɑnsikuentli ui kuld nɑt asɥrteen ɧi
miiniŋ ɑv meni uɥrds ; ɧi distinkիɥn, tu, bituiin
uɥrds ɑv difɥrent miiniŋ and similar sɑund uuld bi
iusles, ɥnles ui liviŋ rɥiters pɥbliի nu iidiիɥns.
In իɑrt ɥi biliiv ui mɥst let piipil spel ɑn in
ɧeer old ue, and (az ui fɥind it iisiiest) du
ɧi seem ɑurselves. With ease and with sincerity
I can, in the old way, subscribe myself,
Dear Sir, Your faithful and affectionate
servant, M. S.”
Speaking of subscribing…
actually, never mind.
Only 11 of the 112 words remained untouched
by the reform, and some of those were just
the inconsistencies we already mentioned.
Implementing this reform would require
completely relearning the spellings of over
90% of English words. It certainly would
have had a better chance in the Age of
Enlightenment than it would today, but would
have required far more backing than it received.
Benjy’s Alphabet is a bit of a mixed bag for
me. It does have some good ideas: the mostly
phonetic orthography, the addition of letters
to tackle some of English’s defective spelling,
and we can’t forget that it’s the only
a priori spelling reform so far to have
addressed morphology. Nevertheless, the illogical
vowel grapheme choices, the inconsistent spelling,
and the sheer amount of relearning required to use
this reform are simply too problematic to ignore.
I guess the “too long; didn’t read” of this
review is: Benjamin Franklin was probably better
at founding America than he was at reforming
English spelling, but hey! That’s not so bad!
All in all, I like Benjamin Franklin’s Phonetic
Alphabet more than I like Parallel English,
but less than I like Cut Spelling, making it my
third favourite global language reform so far.
Thanks for watching, and I’ll see you
next time, where I’ll be reviewing the
Apandah Reform.
[BLOOPER] […] it certainly would’ve had a better chance in the Age of Enlightement- /l~n/a- then it
[clears throat] minemenimanenimenamaah…
Sorry!
© Casual Conlanger 2022