The whole story…

Sep 14, 2011 by

In several previous postings, I discussed the derivation of the Old Novgorod form KѢLЪ and its Modern Russian cognate CELYJ, especially in connection with the first consonant (/k/ vs/ /ts/). But what about its English cognate, whole with an /h/? How did we get this form?

The English word whole is indeed a cognate of the Slavic forms we’ve discussed. So far, we’ve traced the Slavic forms to the reconstructed Proto-Slavic form *kajlu ‘whole, healthy’. Despite what you might think based on the spelling wh, the word whole does not derive from an earlier form that had a hw- or hv- in the beginning, as is the case with many other wh-words in English. For example, what derived from the Old English hwæt and whale — from the Old English hwæl (compare with the Norwegian cognates hva and kval, respectively). The spelling of whole with wh- developed in the early 15th century.

The etymological source of the adjective whole is the Old English form hal meaning ‘entire, unhurt, healthy’, which can be traced, in turn, to a reconstructed Proto-Germanic *khailaz meaning ‘undamaged’. Cognate words from other Germanic languages — all deriving from the same *khailaz — include the Old Saxon hel, Old Norse heill, Old Frisian hal, Middle Dutch hiel, Modern Dutch heel, Old High German and modern German heil meaning ‘salvation, welfare’.

If we dig beyond the Germanic roots of the English whole, we can trace it to the Proto-Indo-European source, reconstructed alternatively as *koylo- or *koilas.

As you can see, it was on the way from Proto-Indo-European to Proto-Germanic that the initial consonant changed from /k/ to /x/, whereas in later Germanic languages it further changed into /h/. This change is a perfect example of lenition, a phonological term that describes a change from a “strong” to a “weaker” sound. In this case, the initial change is an instance of spirantization, a change from a stop consonant to a fricative, whereas the later change is an instance of debuccalization, that is loss of oral place of articulation, from a velar /x/ to a glottal /h/.

This change is part of a larger chain of phonological changes in the history of Germanic languages known as the First Germanic Sound Shift, or Grimm’s Law, after Jacob Grimm (1785 –1863), of the Brothers Grimm fame, who in 1822 elaborated on the earlier discovery of this law by Rasmus Rask. This law describes the development of the stops inherited from Proto-Indo-European into Proto-Germanic in the 1st millennium BCE.

This law consists of three parts. The first part concerns the development of inherited Proto-Indo-European (voiceless) stops into fricatives, of which our story of whole is part. According to part 1 of Grimm’s Law, voiceless stops /p, t, k/ became fricatives /f, θ, x/, respectively (recall that the /x/ later changed to /h/). Since Latin is not a member of the Germanic family (and does not derive from Proto-Germanic), it can be used to illustrate the inherited original consonants. Thus, the Latin /p/ in pēs, pedis corresponds to the English /f/ in foot; the Latin /t/ in tertius corresponds to the English /θ/ in third; and the Latin /k/ in canis corresponds to the English /h/ in hound.

The second part of the Grimm’s Law concerns the development of inherited Proto-Indo-European voiced stops into voiceless stops. Thus, voiced stops /b, d, g/ became the voiceless /p, t, k/ respectively. Once again, this development can be illustrated with the Latin-English contrasts: the Latin /b/, as in verber corresponds to the English /p/ in warp; the Latin /d/ in decem corresponds to the English /t/ in ten; and the Latin /g/ in gelū corresponds to the English /k/ in cold.

Finally, the third part of the Grimm’s Law concerns the development of inherited Proto-Indo-European aspirated voiced stops into plain/unaspirated counterparts. Thus, voiced stops /bh, dh, gh/ became the voiced /b, d, g/, respectively. This development can be illustrated with the following contrasts: the Sanskrit /bh/ in vbhrātā corresponds to the English /b/ in brother; the Sanskrit /dh/ in vidhavā corresponds to the English /d/ in widow; and the Proto-Indo-European /gh/ in the reconstructed form *ghrem corresponds to the English /g/ in grim.

Well, now you know the whole story…

Old Saxon is a West Germanic language, the earliest written form of Low German, spoken c.700-c.1100.

Old Frisian is a language akin to English spoken on the North Sea coast of modern Netherlands and Germany before 1500.

Middle Dutch is the Dutch language as it was spoken and written c.1100-c.1500.

Old High German is the ancestor of the modern literary German language, spoken in the upland regions of Germany; German language as written and spoken from the earliest period to c.1100.

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below:

  • John Cowan

    Whole has some other relatives in English. Hale is a doublet from Scots or Northern English, where OE ā did not > ō, but instead merged with a. Thus in the north āc, bān, rād underwent the Great Vowel Shift, giving MSc aik, bane, raid for E oak, bone, (in)road. (English then borrowed raid in the sense 'incursion' beside native road.)

    Heal on the other hand is < OE hǣlan < pre-OE hailjan 'cause to be whole'. Adding the suffix < (by Grimm's Law) PIE *-itā, L -itas that makes abstract state nouns gives health.

  • Asya Pereltsvaig

    @John Cowan: Thank you for your detailed comment!

  • Seumas

    In Gàidhlig (Scottish Gaelic), you find sounds that are somewhere between the English voiced stops b, d, g and unvoiced p, t, k, so it offers a living example of the transition from voiced to unvoiced. The rules of lenition in Gàidhlig (initial consonant mutation) also show similar unaspirated/aspirated shifts set out in Grimm's Law (p > ph [f], b > bh [v]).

  • Asya Pereltsvaig

    @Seumas: Thank you for your comment! It is true that similar changes (although not the whole set of three) are found in the histories of many languages, and even synchronically in some languages. In fact, lenition is one of the most common types of sound change. As for the "sounds that are somewhere between the English voiced stops b, d, g and unvoiced p, t, k", I am not sure what you mean. I've never heard of sounds that are half voiced and half unvoiced (unless you talk about aspiration, but I don't think you do). In fact, it is a very interesting property of human sound perception that we "hear" such gradual property as voicing as an "all or nothing", either voiced or voiceless.

  • John Cowan

    It's about voice onset timing. English "voiceless" stops are voiceless aspirated; "voiced" stops vary from voiceless unaspirated to voiced. In ScG, the "voiceless" stops are voiceless and intensely aspirated, whereas "voiced" stops are voiceless unaspirated.

  • Asya Pereltsvaig

    @John Cowan: Thank you for your clarification. I don't think you are right on the English stops, as you conflate two different features: voicing (voiced or voiceless) and aspiration (or VOT). English stops can be voiced or voiceless and the voiceless ones can be aspirated (but not always are: e.g., the "p" in "Pat" is, but that "p" in "spat" isn't aspirated). In English voiced stops are never aspirated, but there are languages in which they are or can be (e.g., Sanskrit). As for ScG, the distinction relevant here is aspiration, not voicing. While in terms of articulatory phonetics, voicing and aspiration are related (they are both about the workings of the vocal folds), in phonology the two are typically treated separately. In terms of auditory phonetics too, curiously, we perceive an aspirated "p" still as a "p", not as a more voiceless sound and similarly aspirated voiced sounds (e.g., bh) as an aspirated voiced sound (still a "b"), not a 50-50 or whatever proportion mixed voiced-voiceless (or "50% p and 50% b") — and this is true regardless of which sounds our language actually uses.

  • John Cowan

    My experience is quite otherwise. When I play or pronounce the sequence [pʼa pʰa pa ba ɓa] to speakers of American English with no training in phonetics, and ask them whether they hear /pa/, /ba/, or something else, they uniformly respond /pa pa ba ba ba/, showing that the unaspirated voiceless phone is interpreted as the voiced phoneme. (Scrambling up the order changes nothing. I also add [ʘa] to this sequence, and am always told that this is neither /pa/ nor /ba/.) What is more, whispered speech maintains the distinction between /pa/ and /ba/, where the only phonetic difference is the aspiration.

    On this view, the phonetic quality of /sp st sk/ is neatly explained as a neutralization of (phonological) voicing rather than a loss of (phonetic) aspiration, though of course both statements are true. The Celtic languages also neutralize stop voicing after /s/. In ScG and Welsh these clusters are written sb st sg; Irish used to use this convention also, but has reformed the spelling to sp st sc, probably after English. I suspect the strong aspiration combined with /s/-neutralization is a British Isles Sprachbund effect.

  • Asya Pereltsvaig

    @John Cowan: if your experimental experience is different from that replicated and published by others, how do you explain that difference? Since I am neither a phoneticist nor a phonologist, I haven't conducted similar experiments myself, and have no explanation of why you get unusual results.

  • Seumas

    I'm not a linguist so I can't add to this other than to say I know what John Cowan is referring to. In Gàidhlig, b-p, d-t and g-k/c sounds all exist the same as in English but there is an intermediate sound for each that doesn't exist in English which English speakers mistake for the 'full' sound. For example:

    gorm – the 'g' is hard like in English
    agam – the 'g' is like a palatised 'k' half-way between 'g' and 'k', English speakers hear this as 'k' but it's not.
    ceann – the 'c' is like English 'k'.

    bean – 'b' as in English
    Alba – a soft 'b' which English speakers hear as 'p' but it's not.
    peint – same as English 'p'.

    In any event they give English speakers almost as much pain as the 'gh/dh' palatised sounds. Whether it's aspiration or voicing I don't know, but to me they sound like a continuum.

  • Asya Pereltsvaig

    @Seumas: perhaps the perception of continuum on the part of ScG speakers has more to do with how they are taught about these sounds, how they are presented in grammars than with the actual articulatory and acoustic properties of these sounds. As you mention, English speakers don't perceive them as a continuum, but as either [g] or [k]…

  • finnish website translation

    Its Great Story….. thanks for sharing. this is amazing blog.

    finnish website translation

  • Asya Pereltsvaig

    @finnish website translation: thank you! keep reading!

  • Pingback: Palatalization in Old Church Slavonic - Languages Of The World | Languages Of The World()