Episode 5: Centum, Satem and the Letter C

A look at the early division of the Indo-European languages into the Centum and Satem languages.  The sound shift which marks the division of the Centum and Satem languages is then explored in the context of the modern English letter ‘C’. The history of the letter C is presented from its Greek origins to its modern usage.

TRANSCRIPT: EPISODE 5

65 thoughts on “Episode 5: Centum, Satem and the Letter C

  1. The Old English word cyning “king” is not pronounced /kining/ but [ˈkyniŋɡ]. The vowel as it still exists in today’s French mur or German müde.

    • You are correct about the pronunciation of ‘king.’ At the time I prepared this episode, I wasn’t particularly concerned about the specific pronunciations of the Old English vowels (and I mentioned that fact in a couple of episodes). However, as I started to transition into the Old English period around episode 27, I became more concerned about the pronunciation of Old English. So I think you will find that the pronunciation is a bit more accurate once you get to those episodes. (Not perfect, just a bit more accurate.)

      • Discovered the podcast from a colleague who kept mentioning your work in conversation. He gave me the first few episodes and now can’t stop listening to them. It’s absolute brilliant work thank you so much. I’ve got a question related to the thread above, apologies if it’s naive or answered elsewhere but how do we know how words or syllables were pronounced orally, as per your interaction with Hanspeter Herbash, when the language is dead and only written resources are available?

        Thank you again and fantastic work

        • The issue of pronunciation is actually discussed in future episodes, but in general, most medieval writing was phonetic because there were no dictionaries or standard spellings. So scribes wrote down the words the way they were pronounced. Very often, scribes in various parts of the country spelled words differently due to different dialects, but within each region the spellings were somewhat consistent. It is also important to keep in mind that each letter of the alphabet represented a specific sound in much of the Middle Ages. Today, letters can represent lots of different sounds, but that’s because spellings were standardized several centuries ago and the pronunciations continued to evolve. Thus, modern spellings no longer reflect the actual pronunciation of words in many cases. But in the Middle Ages, the spellings would have more accurately reflected the pronunciation.

    • Except that spellings such as cining, cincg, cing, cyneg, cyng, cyncg, cyneg, all existed, so the pronunciation of the “y” could have been /y/ or /i/ or differed by dialect or scribe. Check Bosworth-Toller.

  2. When the Fijian language was being written down, several letters of the alphabet were reassigned. D became ND as in Nadi and B became MB as in Bula. Biblical names are modified, David becomes Tevita and Elisabeth becomes Lisapeti. G sounds like the NG in singer and Q like the NG in finger. The island of Gau sounds like Now. The drink which the Polynesians call Kava is called Yaqona by the Fijians, But most significantly C sounds like the TH in the. There is no F or J in Fijian. The homeland is VITI.

  3. Kevin, this is a wonderful podcast. Thank you. I look forward to continuing. The question that comes to mind when listening to this podcast is that of ‘ck’ and where that fits. Is this covered in a bonus episode of answering our questions. If so, I wait with baited breath.
    Thanks again,
    Anna

    • Hi Anna. I touch a little bit on the development of the ‘C-K’ spelling in “Episode 89: ‘I Before E’ and All That.” I discuss the evolution of that spelling the context of short vowel sounds.

  4. Kevin, by showing how root words coming from the proto-Indo-European language gave rise to cognates in its descendents which aren’t readily apparent until one is aware of the patterns of changes in pronunciation of key consonants, you’ve made seeing the parallels in each language family so much easier. Thank you for sharing this knowledge! I look forward to what I’ll learn listening to all your episodes. In the first six, I’ve found answers to language questions I’ve had for years.

  5. It might look funny to say I’ve heard 6 episodes in a comment for #5, but I was re-listening to #5 when I wrote my previous comment!

  6. Pingback: Cider By Any Other Letters Spells As Sweet | Pommel Cyder

  7. Hello Kevin,

    Big thanks for creating these superb podcasts – full of wonderful information! So many rich insights into what shaped English, its alphabet and its irregularities that I have wondered about..

    In this episode (5:20), you said, the work for hundred in Sanskrit is “satam”. The correct word is “shatam”.

    • Thanks for the note. It’s funny because many of my sources say that “satem” is the Sanskrit form of the word, but upon further research, you are correct. It is the Avestan form of the word.

      • Telugu is credited for preserving a most of Sanskrit be sounds. In Telugu the word is satum with a soft ‘t’ and the u pronounced as the shorter ‘a’ as in the word Minnesota.

        Hindi the alphabet used for s in satum is श. But श and ष make same sound in modern hindi. This I believe is a corruption of the alphabet श. I think that is the confusion here.

        It is explained here a bit better.

        https://m.youtube.com/watch?v=woGYr3EtuB4

  8. I love your podcast! Incredibly interesting. Thank you for sharing your wisdom, passion.
    I’m curious about “ocean” … the sh sounding “c”…

    • I’m not sure about the linguistic history of “ocean.” It was borrowed from French in the early 1300s. I think it is probably related to the same development that occurred with words like “sure” and “sugar” in the Middle English period. In all of those cases, the ‘s’ sound became an ‘sh’ sound before the vowel.

  9. Being very much the layman linguistically and history wise, I had thought there was no k in latin. But if they got it from the greeks how did the greeks differentiate the sounds and usage?

    Really good cast by the way—coming from THoR, Byzantium, Crowther’s England & Jeffers’ British History.

    • The short answer is the Greeks used the original version of C (which was gamma) for the /g/ sound and the original version of K (which was kappa) for the /k/ sound. The sounds became confused in the Etruscan language which didn’t have a /g/ sound, but which had several different variations of the /k/ sound. So the Etruscans kept the K, and they changed the C from a /g/ sound to another one of their various /k/ sounds. (The /k/ and /g/ sounds are essentially the same except that the former is voiceless and latter is voiced.) A short time later, the Romans borrowed the Etruscan version of the Greek alphabet. Since Latin only had one basic /k/ sound, the Romans couldn’t really distinguish the way the Etruscans used the letters C and K. To them, each letter had the same sound. Since they didn’t need two letters for the same sound, they tended to just use the letter C. The K is occasionally found in Latin texts and inscriptions, but it is very rare.

  10. Very interesting episode.

    It reminds of when I was in Norway a few years ago and was surprised to discover that ‘K’ was sometimes pronounced with a sound similar to “ch” in English.

    I met someone called Kirsten whose name sounded (to me) like ‘chur-sten’ when people were saying it. However, the K was pronounced “hard” in other words such as:

    beklager sorry
    kvittering receipt

    This episode also helped me make sense of the fact that we still have the word ‘kirk’ in the UK referring to a church, particularly in Scotland.

  11. Hello from New Zealand and thanks for a great series, loving these podcasts. I don’t know much about language, I only speak English and am a terrible speller but have enjoyed learning more and really appreciate the history lesson. Thanks so much again, Rachael.

  12. Hi, Kevin. Really interesting podcast thank you. Any connection there or reason for the development of the ‘ck’ grapheme? You also mentioned links to ‘ch’ as a variation but is there a story for ‘sh’ or ‘x’ phonemes?

    • Hi Kelly. I touch a little bit on the development of the ‘ck’ spelling in “Episode 89: ‘I Before E’ and All That.” I discuss the evolution of that spelling in the context of short vowel sounds. I’m not sure about the second part of your question, but the /sh/ sound emerged out of the /ch/ sound in French, thereby producing the modern pronunciation of words like “chef” and “champagne.”

  13. After listening this morning to episode 5 konserning the letter C, I stand firmly on the side of the abolitionists. C is a krippling vestige of our early lingual ansestors’ subjugation under the mersiless Frentsh, and serves no purpose in modern English writing and dialogue. I suggest that we embrase our Germanik roots with pride and resist assibilation into this Frentsh monstrosity.

    #abolishC
    #resistassibilation

    Exsellent podkast, by the way.

    • How will you distinguish between sell and sell then? You need C to distinguish between words like sell and cell (cent Vs sent etc)

  14. Which brand of Spanish are we talking about?
    I’d suggest History of Spanish Consonants site and The History of Spanish, A Student’s Introduction by Diana L. Ranson, Margaret Lubbers Quesada, Cambridge University Press, 2018. “Ciento” is pronounced “thee-en-to” in Spain but “see-en-to” in Andalusia (but not everywhere there) and in Latin America. The pronunciation of “ce”-“ci” wasn’t uniform in all the Spanish speaking world in the 17th century either.

  15. This episode made me think of the clever French way to turn a potentially soft “c” into a “k” when the following vowel sound would normally dictate soft: the cédille. One can imagine some scribe coming up with this notation on her/his own to avoid confusion.

    Thanks so much for your wonderful work! I have just started listening and am hooked!

  16. A question:

    Given the satem/centum split between the various branches of the Indo-European languages, how did linguists know to reconstruct the sound as ‘k’ in the Proto-Indo-European language?

    I find this particularly puzzling given that the home of the original language is located right in the middle of the modern day Satem languages.

    What clues did they have to suggest that it was the ‘s’ sound that was acquired. Perhaps it is known that the shift has never been observed to go the other way, but I don’t find that to be a rigorous proof. It is the kind of ‘proof’ used by climate change deniers: “Climate change has only occurred before without a human cause, so it can’t be humans who are causing the change this time”.

    Is there some other evidence which you haven’t mentioned?

    • Hi Bob.

      Great question, and frankly, it is a question that would be better answered by a proper historical linguist. I am mostly presenting the conclusion of scholars who have studied this issue thoroughly. My understanding is that the sound was reconstructed as a ‘k’ sound using the generally-accepted sound changes that are thought to have occurred within the various Indo-European languages. More specifically, I think the ‘k’ sound within PIE is supported by the fact that the Tocharian language of western China was a Centum language. The Anatolian languages (including Hittite) may also have been Centum languages, and they are thought to represent the oldest branch of the Indo-European family tree. However, the Anatolian branch is more problematical as there is conflicting evidence regarding its classification. Some linguists even contend that the Anatolian branch represents a pre-PIE offshoot that emerged slightly before PIE was spoken in its modern reconstructed form. At any rate, the existence of the ‘k’ sound within those early branches provides some evidence to support the idea that the original sound was a ‘k’ sound.

      • Thanks Kevin for the reply.

        There’s another issue which muddies the waters for me.
        The Uralic language family is posited as the most likely sister family to Indo-European. Words for “hundred” in some Uralic languages are:
        Hungarian: száz
        Finnish: sata
        Estonian: sada
        Udmurt: сю (where the ‘c’ is pronounced as an ‘s’)
        Northern Sami: čuođi (where the ‘č’ is pronounced ‘tsh’)
        Nenets: сотня (where the ‘c’ is described as a ‘voiceless fricative’)

        The Finnish word suggests the connection to Indo-European. And the fact that I couldn’t find a single ‘k’ word for ‘hundred’ suggests to me that Uralic languages could be classified as Satem languages. So either the shift from ‘k’ to ‘s’ really is a near-certainty (in which case you have to ask why the centum languages were so resistant to the change), or the concestor of PIE and Proto-Uralic (and PIE itself??) is actually a Satem language.

        Do you know of any expert who I could raise this with?

        • I don’t have a specific expert to recommend, and unfortunately, I don’t have any reliable etymology resources for the Uralic languages. For what its worth, wiktionary.com says that the examples you cited represent an early Uralic word that was borrowed from a Satem branch of PIE. It specifically says that the word was borrowed from the Proto-Indo-Iranian branch (which is a Satem branch). If true, that would account for the ‘s’ sound in those Uralic words. I wouldn’t consider wiktionary.com to be a 100% reliable resource, but it is the only information I have.

        • I’ve read some works by Hungarian linguists (nothing too serious or hardcore, just university lecture notes and a book meant for the wider public – but perhaps still more reliable than wiktionary). It seems to be consensual among them that the word for “hundred” was borrowed from a satem Indo-European language (most likely some Iranian language) in Proto-Finno-Ugric, and then inherited by Finnish, Hungarian etc. There are some other examples of Proto-Finno-Ugric contact with Indo-European languages, and Hungarian continued to borrow from them for a while after becoming a separate language.

  17. I’ve just discovered your brilliant podcast. I’m a language nerd (one time Latin and Greek scholar). Just wanted to point out that Scottish and some northern English dialects use “kirk” for “church”.

    • Hi Nigel. Yes, the “kirk/church” example comes up quite a bit in the later episodes. It’s a classic example of the Norse influence in northern England, and it also illustrates how northern English dialects often retain a ‘k’ sound where southern dialects have a softer ‘ch’ sound.

  18. I just started this series this week, right after finishing your Beowulf Deconstructed. I was giddy when I got home and told my wife everything I had learned from this episode. I’m glad to know there is so much more to come!

  19. Kevin, I discovered your podcast last week and am loving it! I’ve had some real lightbulb moments so far, features of the English language that I had never thought of. I just wanted to say how much I am enjoying the series, thank you. Myself, I teach classical Greek and send subscribers a newsletter weekly, and will feature the podcast in this week’s edition. Best wishes, Helen

  20. I am a native Russian and Ukrainian speaker, and a comparative linguist by training. Listening to your wonderful podcast sends me right onto the cloud nine where I firmly and happily remain while engaged in matching, collation, contrast, similarity between the Germanic, Romance, and Slavic languages, their sounds and spelling. So many similarities! So many interwoven threads and connections throughout history and tribal movements.
    Thank you very much! Absolutely delightful and enlightening lectures.

  21. How very fascinating.
    As a native speaker of Hindi language (belonging to the Sanskrit family) , I sometimes wondered if there was any connection between the set of words meant to ask questions such as “kyun”,”kya”,”kab”, “kaise”, “kahan” in Hindi meaning “why”, “what”, “when”, “how”, “where” respectively and if it has anything to do with K and H sound pair. It gets more and more interesting.

    English has a fascinating story.

    Thank you for sharing with us.
    Sincerely,
    Shubo

  22. Hi Kevin,

    A linguistics student mentioned this podcast to me the other day and I’m really enjoying it! Thank you so much for all the work you’ve put into this. You have a really good pace, you anticipate and answer the questions in my mind as you go along, and you draw in parallels that help with understanding the points you make, as well as recapping periodically to help it sink in. Excellent stuff!

    Can I ask about the first letter c in Celtic? It’s pronounced as an s with the Scottish football team Celtic Rangers, but the Celts have a hard c.

    Chris

    • You have a very interesting article about it on the Merriam-Webster site but in a few words, it was borrowed from the 16th century French so with /s/ but in the 18th century academics started to pronounce it with /k/ to reflect the original ancient Greek Keltoi and Classical Latin Celtae pronunciation.

  23. Thank God I lived to find find this podcast! I understand twelve Indo-European languages to some extent, and now I can explain the similarities I’ve always felt. Never heard anything as fascinating. Of course, I’d like to add a couple more members of this familiy to my list, especially Hindi and Farsi, but after listening about all the influence Greece and Greek had on the word, I feel like I cannot let that piece of cake pass. Of course Italian (Latin) will never be underappreciated 😉 Thank you so much for providing this info! Awesome work!

  24. Very interesting. Only five episodes in, but I think I’m hooked!

    Look, I know this isn’t ‘The History of French Podcast’, but based on my knowledge of French I’m still somewhat confused about where the English letter K came from. Towards the end of this podcast, you said the Middle English scribes copied the tendency of French scribes to render a hard /k/ sound before a front vowel using the letter K. Yet, the K hardly appears in modern French at all, pretty much only in loanwords. What you do see is the /k/ sound being written Qu before a front vowel, as in “qui”, “que”, “quel” and “question”. I’m no linguist, but the idea that a single sound, /k/, underwent two spelling changes in the same language over a period of a few hundred years (from C to K, then from K to Qu) stretches my credulity. Do you know if there was in fact a spelling change in French from K to Qu, or if perhaps the English letter K as in “kind” and “keen” was copied from a language other than French?

      • Merci Monique, très intéressant. It’s nice to imagine 18th century French people getting excited by “exotic” K-words from other languages.

    • Hi James. I’m glad you’re enjoying the podcast. With respect to your comment and question, I don’t think there were two different spelling changes as much as there was inconsistency in the way words were spelled with some scribes using the largely forgotten letter K in certain situations. As noted in the episode, the letter K was rarely used in Latin (pretty much limited to loanwords), and it continued to be rarely used in early French. Apparently some French dialects preferred it more than others, and some used it to deal with the assibilation issue I described in this episode. Middle English scribes adopted the letter because they knew that it was sometimes used (albeit rarely) to represent the /k/ sound in Latin and French. At any rate, I thought it might be helpful to include a quote from the Oxford English Dictionary’s discussion of the history of the letter K:

      “To Latin scribes of the sixth, seventh, and eighth centuries, K was thus known as a supplementary letter to C, of use in Greek or other foreign words which had the ‘hard’ or k sound of C before e, i, or y. Hence it was naturally put to use in the writing of Old High German, Old Saxon, Old Frankish, Early Italian, and some dialects of Old French, in which a k sound came before e, i, or y. In writing these languages, C was usually employed, as in Latin, before a, o, u, or finally; but in practice there was considerable overlapping, with the final result that, in German, K ousted C, and is now the proper letter for this sound in that language, as well as in Dutch and all the Scandinavian tongues; while, in French, K was ousted partly by C, partly by Qu, according to derivation. (Thus Old Northern French kanon, karole, katre, ke, ki, kel, became later canon, carole, quatre, que, qui, quel.)”

      • Sorry, I forgot I asked this so didn’t see your very informative answer until now. Thank you for that and for the brilliant podcast series (I’m up to E21 now).

  25. Fascinating!! I never would have imagined that the history of a single letter could possibly be interesting. Like James, just a few episodes in and I am hooked! Thank you for this podcast. I very much appreciate the way you’ve structured the podcast and broken down this enormous topic – it’s perfect for the layperson who has always been fascinated by languages and human history but never studied them. Well done!

  26. Brilliant podcast!
    A minor comment about Latin pronunciation. Very sensibly, you use the reconstructed classical pronunciation.
    It might have been worth pointing out this is different from the ecclesiastical pronunciation. The majority of people who studied Latin at school in a country with a catholic background, only use the latter and, I know for sure, are unaware of the reconstructed pronunciation.

  27. I’ve just found this podcast. It’s wonderful. Thank you so much. It’s fascinating to learn about how English developed. Very useful for me when teaching English as a foreign language. Thank you for your scholarship

  28. Thank you for this podcast, it’s incredible how much detail you can cast light on within the smallest topics.. Much appreciated!

  29. Beautiful podcast. Thank you so much.
    It makes sense why we say in Spanish:
    ca, co, cu = K
    ce, ci = S
    que, qui = K
    I have been a Spanish teacher for 26 years and I love knowing the origin of words. Thank again!

  30. I am finding everything fascinating. Thank you for sharing, it has opened my world up!

    I would love it it if you could create a child friendly version, language tree, main changes to English, similar to the story of C, you tell it so much better than I ever will!

    I love the way your information is delivered. People coming in at any level could take away new knowledge. History in any form helps us understand the world we are in.

    • Hi Allyson. Thanks for the feedback. If I can find the time, I would like to produce a companion series in which each episode focuses on one aspect of English that people find challenging and attempts to explain the historical basis behind it. It is difficult to fit it into my schedule, but stay tuned! Maybe I’ll be able to produce it in the future.

  31. I am interested in ‘concern’ vs ‘concur’.
    These days ‘er’ and ‘ur’ are pronounced the same, at least where I live. Does the fact that one is preceded by an ‘s’ sound and the other is preceded by a ‘k’ sound mean that ‘er’ and ‘ur’ had distinct pronunciations in the past?

    • “Concern” comes from middle French (“concerner”) itself from Latin and “concur” comes from Latin (concurrere). These words entered the English language in the early 15th century, long after the ce/ci > se/si assibilation happened. So whether they were pronounced the same as now or not in English doesn’t change anything as the spelling was borrowed along with the words.

  32. Thank you very much for this podcast. I just started listening.

    I wonder if linguistics offer an explanation why the kentum/satem split appears not to follow the tendency of assibilation of k in front of e and i (not a as in satem).

  33. The reason why the grapheme “c” represented the [k] sound in some cases instead of the [s] sound might be popular evolution, that is, speakers of the language, unconsciously, would interrupt the “natural” phonetic evolution of a word because of the specific meaning of said word. The example given is the word “king”. Perhaps people —again, unconsciously— wanted to keep the word “king” with the old sound [k] as a sign of respect towards … the king. Or perhaps the logical resulting word *sing could be confused with a homonym word (the verb “to sing”?), another non-desirable result.
    Something similar happened to the name of the Spanish city Mérida. Originally EMERITA in Latin, it would have naturally evolved into *mierda (Spanish ‘shit’). Obviously, people wanted to avoid such scatological result, so they interrupted the natural evolution of the word.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.