I just ran across the Finnish word lounas, which means both ‘lunch’ and ‘southwest’ (this was in the context of the former dacha community Lounatjoki northwest of St. Petersburg, since 1948 called Zakhodskoe; there is some dispute over whether the name originally meant ‘southwest river’ or ‘lunch river,’ the argument for the latter being that workers ate meals there while they were building the railroad that was the community’s raison d’être). On a whim (the kind of whim that comes naturally to me) I looked up “southwest” in my Estonian dictionary to see what the sound correspondences looked like, and was mildly disappointed to see it was an entirely different word, edel. But wait: just above it, under the rubric “south,” was lõuna! I looked up “south” in the Finnish dictionary, and sure enough, it was etelä. It seemed odd that the two words would have exchanged senses; a little googling got me this passage from p. 216 of Basic Aspects of Language in Human Relations: Toward a General Theoretical Framework, by Harald Haarmann (Mouton de Gruyter, 1991):

The kind of “oscillation” in the meaning of basic terms which has been illustrated for the Lappish terminology finds its equivalent when comparing other Finno-Ugric languages. As an example, I refer here to contrasts in the corresponding Finnish and Estonian terminology: [Here he gives a table with the above words.] These and other variations are, as in the case of Lappish terms, a reflection of orientation according to weather conditions and the profile of the surrounding landscape, rather than an orientation in terms of the position of stars in the sky.

I thought that was interesting, so I’m passing it along. (The book sounds interesting; anybody know anything about Haarmann?)


  1. Sometimes cardinal directions are named astronomically (for example as sunrise / sunset / noon / midnight), and sometimes, meteorologically (by wind directions). In the latter case, 45 degree variations are not uncommon.
    For example in the Romance Mediterranean, Libeccio is a SW wind (<= from the direction of Libya / Africa i.e. also kind of South rather than SW).
    Sirocco is the South-Easterly wind but its name is derived from the Arabic word for “East”.
    Bora wind, related to Boreal “Northerly”, changes direction from coast to coast (it’s more like whatever blows from the nearby highlands into the sea). A similar but reversed situation is with Tramontane, literally a wind from across the mountains but used for Northerly winds.

  2. BTW the Pomors of the White Sea Coast, who had significant Finnic cultural influences, used Russian adjectives for Winter and Summer to designate the cardinal directions of North and South. AFAICT this pattern is not used in modern Finnish or Estonian. Does anyone know what is the origin of the Pomor usage, and how common it may be elsewhere in the world?

  3. Sorry for the continuing piecemiel spam 🙂 I was wondering how the Finnic people may have become accustomed to the wind notations if they weren’t grand old maritime nations, but I guess I got an answer.
    The huge lakes of the Karelo-Finnic heartland have very severe winds, and some local wind names persisted far beyond the lakes of their origin. For example Shelonnik (шелоник, шалонник) is named after the Shelon’ River in the SW corner of Lake Ilmen’. There, the SW wind used to endanger the the trade vessels of Novgorod merchants. The name survived for centures after the downfall of Great Novgorod in the language of, yes, the Pomors, where it continued to designate South-West.
    BTW the Pomor designation for South-East is … yes, literally Lunchtime (obednik). Here we go with another 45-degree shift from lounas “south or SW or lunchtime” – or SE as well.

  4. I would put my money on ‘south river’.
    /lounas/ (fin), /lõuna/ (est), /lounat/ (kar): ‘south’, ‘southwest’, ‘midday, ‘lunch’ (which happens to be the meal eaten while the sun is in the south!)
    /lun/ (udm), /lun/ (kom) ‘day’
    So the meaning ‘lunch’ has been derived from the ‘day’, ‘south’, ‘sun’ (which tend to overlap in many Finnougric languages). So no coincidental homonymy there.
    As for the village name: surprise, surprise: there runs a river through the village which goes by the name of Lounatjoki (at least back in the days when there were Finnish-speaking people there. As names of rivers tend to be older than names of settlements, as the river probably was there before the village, I guess the village was named after the river.
    Other toponyms in the area: Hunger swamp, Roach lake, Great forest swamp, just to give an idea of the geography and how well South river would fit in.
    Source: Suomen kielen etymologinen sanakirja, map of the Karelian Isthmus. I’m a linguist in Helsinki.

  5. the meaning ‘lunch’ has been derived from the ‘day’
    In Old English the word for lunch was also nonmete <= “noon”. And Russian “obed” (lunch) is etymologically just “a big meal”. There also exists a different Russian meal-time name, полдник poldnik “noontimer”, and resp. adjective полуденный may mean both “midday” and “southerly”. But of course in contemporary usage, Russian poldnik is just a light snack esp. for preschoolers, rather than lunch.

  6. I’m a linguist in Helsinki.
    Then I hereby declare your opinion valid!
    *bangs gavel*

  7. Trond Engen says:

    Time of day as cardinal direction is straightforward. There are plenty of mountains and hills in Norway named Middagsåsen, Dugurdstind, Nonsknatten etc. Named for rheir use as a sundial cum meal-marker on some habitated spot.
    I think I’ve told before that in the valleys of Eastern Norway, the tradional system of directions had north up the valley. In some sidevalleys of sidevalleys, the sun rose in the west and set in the east, and the midday sun was due north.
    While I’m at it, I may also have told that I don’t think the roots of Norw. øse “empty of water” and øst “east” are accidental homonyms. I like to think the PIE word for east meant “unleashing”.

  8. Trond, about øse, I’m sure you know the River Ouse in northern England:

    The word ‘ouse’ is a very common name for rivers in England – it derives from the Celtic word ‘Usa’, from *udso-, which simply means ‘water’. ‘River Ouse’ therefore actually means ‘River Water’, etymologically.

    It has been suggested that the ‘Ouse’ was once all known as the ‘Ure’, but there seems to be no supporting evidence for this claim. In fact, more credence is given to the assertion that the name derives from the Old Celtic word for ‘Ure’, ‘Isara’, which over time evolved into ‘Isure’, ‘Isurium’, ‘Isis’ and finally the Saxon ‘Ouse’. This linguistic evolution also goes some way to explaining how the little tributary ‘Ouse Gill Beck’ which enters at Linton-on-Ouse usurps the name of the much larger river ‘Ure’.

  9. Another split in meaning can be seen in the pair sulane (est) – sulhanen (fin), meaning “farmhand” and “fiancé, groom”, respectively. I’m strongly reminded of the story of Jacob, Rachel, Laban and Leah.

  10. Harald Haarmann is an astonishingly prolific German linguist: he started out as a Romance specialist, writing books on Latin loanwords in Breton, Welsh and Albanian and compiling a wordlist of Cubanisms in the works of Fernando Ortiz, then married a Finn, after which he moved on to minor Uralic languages, wrote a very useful book on Uralic sociolinguistics together with his wife, a thick tome on the influence of Russian on Ingrian, a book on evidentiality as a Eurasian isogloss, and edited reprints of old Estonian grammars. Then in the late 70s (I think) he moved to Japan, and his purview became even wider: his latest publications are more general, and he’s recently written books on multilingualism, writing systems, extinct languages of the world, the Indoeuropeans, the history of numbers, the languages of the world (worth reading, though I’d say Ostler is more engaging), all aimed at a wider audience. I have no idea how he manages…

  11. Birdseeding says:

    Hungarian – in common, it appears, with these other finno-ugric languages – has words for directions and times of day that are closely related:
    Dél (both noon and south)
    Észak (north) — éjszaka (night)
    Kelet (east) — napkelte (sunrise)
    Nyugat (west) — napnyugta (sunset)
    All forming a neat compass-to-day-cycle correlation. Since the sunrise-sunset thing is a fairly obv way of getting east and west, could the other directions somehow derive?

  12. This was probably obvious to everyone here, but just in case anyone missed it: in the Northern Hemisphere, the midday sun is to your south, so there’s a natural association “noon”=”south”. The further north you go, the more obvious this gets, but even France has its Midi region.
    Throughout North Africa, “south” is qiblī. Etymologically, this comes from qiblah – the direction of Mecca – even though Mecca is pretty much due east; but the Arabs who first reached north Africa came in from Palestine and the Hijaz, where Mecca is due south.

  13. Harald Haarmann is an astonishingly prolific German linguist
    Thanks for that comment; I’m glad to know about him!

  14. “The word ‘ouse’ is a very common name for rivers in England – it derives from the Celtic word ‘Usa’, from *udso-, which simply means ‘water’. ‘River Ouse’ therefore actually means ‘River Water’, etymologically.”
    Mongolian word for water is us ( pronounced as in ou too not as ah in we-us) too, just coincidence or some kind of connection?
    i know in Tatar it’s su(from Alsu – rose water how i have learnt that) and Japanese mizu with the zu sillable also seem like close sounding, all might be just coincidences of course or some very distant now language connections, i mean if Celtic is considered in the Finno-Ugric family, maybe it is not, then maybe it’s not surprising how come in the PIE territory of water- voda or aqua suddenly usa for water

  15. There also exists a different Russian meal-time name, полдник poldnik “noontimer”, and resp. adjective полуденный may mean both “midday” and “southerly”.

    Ужин” (supper) was originally a midday meal (< “юг“, south).
    Of note is also Pomor “пАyжна” – a meal after (“по-“) midday. Confusing w.r.t. modern usage of “ужин” as evening meal, since “паyжна” is now before “ужин“. (For example, Pisakhov: “Спал до вечера и паужну проспал. К ужину явился…”)
    And there’s corresponding wind too :), “пАужник” = SW.

  16. Read, I don’t know if there’s any connection, but it would be nice if there were. In English there’s also ooze (to seep liquid). There is a Scottish word oos, but it means dust bunnies, the fluff under the bed (wool is oo to a Scotsperson, apparently).

  17. “Ужин” (supper) was originally a midday meal (
    Interestingly, Russian “poldnik” (midday meal/snack) corresponds to Spanish
    merienda / Tagalog meryenda / Croatian marenda, which is usually considered to be a cognate to “merit” (a meal well deserved, so to say), but not to “meridian” (middle of the day / noon / Southerly direction).
    Wouldn’t one expect a merienda / meridian connection instead?

  18. silly syllable
    wool is noos (pronounced nohs, not like in ooze which is nevchikh) so it seems different, celtic wiki says is from proto-celtic which is a branch of PIE, not finno-ugric/uralic
    but ys is hair/fur, pretty close sounding to oos, but it’s more like umlaut, so

  19. It’s the mi- part that means “water”, not the -zu.
    港 minato = mi-na-to “water’s door”, or simply “harbour/port”
    汀 migiwa = mi-kiwa “water’s edge”
    源 minamoto = mi-na-moto “water’s source”, “fountainhead”
    水底 minasoko = mi-na-soko “water’s bottom”
    水上 minakami = mi-na-kami “water’s top”, ie “the upper reaches (of a river)”
    水面 minamo = mi-na-mo “water’s surface”

  20. what about it’s sui pronunciation then?

  21. sui is an adaptation of the Chinese shuĭ.

  22. ah, onyomi, fine, sinitic languages are surely completely by themselves, and japanese have more direct borrowings from chinese than any that , core but distant connections to uralo- altaic languages, ok

  23. Trond Engen says:

    AJP: I’m sure you know the River Ouse in northern England
    Yes. *ud- of *udsa seems to be the oblique stem of the root *wed-/wod- known from ‘water’. Germanic regularized the paradigm after the nominative, but the oblique stem is preserved in otter &lt *ud-r-ó- (says B&L).
    Norw. øse and øst are monophtongized forms, probably from Danish but also regular in some areas along the Swedish border, so I would have been wiser to write ause and aust. Still, Ouse is unrelated. since *aus- &lt PIE *h2ews-. This diphtong corresponds to Eng. ea- as in east. An English cognate of ause would look exactly like ease, and the semantics is a decent fit too, but it’s unfortunately from French.

  24. marie-lucie says:

    Dmitry: Spanish merienda (hence Tagalog);
    Without more information, I think that it is quite possible that Spanish merienda rather than something like meridiana is due to a folk-etymological reformation, since the initial mer… suggested mérito and merecer while the rest of meridiana had no obvious meaning to ordinary people (even if it was used in scientific contexts, those contexts would be too different from that of meals).
    About the times of meals: In both French and English the words for meals taken during the day (noon and evening) have shifted from their earlier, apparently because of social mores. As the idle rich went to bed later and later (after going to plays, gambling, etc) and consequently got up later and later, the first meal (French le déjeuner, literally ‘to unfast’), ended up referring to the noon meal, and the former noon meal le dîner was taken in the early evening. English kept breakfast as the morning meal taken by all social classes, but dinner (from French) moved later and later (Christmas dinner in the early afternoon is a survival). French le souper which used to be a light evening meal was borrowed as English supper which may or may not be the equivalent of dinner (if you are invited for dinner when away from home, you should always ask what time you should arrive, or you might be much too early or much too late). In French le dîner is considered more elevated than le souper which is more a family term if used for early evening: you could go to a restaurant for dîner at the normal time for this meal but not for souper unless you wanted a light meal after the theatre, for instance, or a midnight meal on New Year’s Eve. (This is the usage in France, other countries might have different usages).

  25. Trond Engen says:

    uwe: “Ужин” (supper) was originally a midday meal (
    Also Norw.
    middag “high noon” is eaten (by most people nowadays) in late afternoon or early evening, traditionally the big meal after daddy’s home from work. I was going to ask if this is a universal, but…
    M-L: About the times of meals: In both French and English the words for meals taken during the day (noon and evening) have shifted from their earlier, apparently because of social mores.
    Similar to French, Danish frokost means “lunch” while “breakfast” is morgenmad.

  26. marie-lucie says:

    the words for meals taken during the day (noon and evening) have shifted from their earlier, apparently because of social mores.
    Oops, in editing this sentence I forgot the word “meaning” after “earlier”.
    I will add that after le déjeuner switched from the meaning ‘breakfast’ to that of ‘lunch’, there was no longer a word for the early meal, which was now referred to as le petit déjeuner ‘the little breakfast’, which, unlike the morning meal in the English tradition, is not a full meal but more of a snack, and consequently, the noon meal is a full meal, often the main meal of the day. (This is changing to a certain extent as working people take/are given less time to eat at noon in exchange for quitting earlier).

  27. marie-lucie says:

    Lameen: in the Northern Hemisphere, the midday sun is to your south, so there’s a natural association “noon”=”south”. The further north you go, the more obvious this gets, but even France has its Midi region.
    France is not the only place: Italy has its Mezzogiorno, transparently ‘Mid-Day’. Occitan (surviving in the French Midi area) has (in the very conservative official spelling) Miègjorn, also ‘Mid-Day’. I am not sure what the official pronunciation is, but I think my grandparents would have said [myedzún].
    I am told that le Midi, which used to refer to the Southern half or at least third of France, is now only used for the Mediterranean area, while the rest of the region is le Sud. I think this may be another anglicism, since le Midi cannot be translated into English except as “the South”, retranslated as le Sud.

  28. I think that it is quite possible that Spanish merienda rather than something like meridiana is due to a folk-etymological reformation, since the initial mer… suggested mérito and merecer while the rest of meridiana had no obvious meaning to ordinary people (even if it was used in scientific contexts, those contexts would be too different from that of meals)
    It may be similar with Russian uzhin dinner which has origin rooted in science-usage and Old Slavonic … and is readily lending itself to a folk etymology linking it with the root uz/uzh “narrow”.
    Meal-names tend to have late-shifts in all societies, which is why I mentioned Tagalog as a curious example of meryenda shifting to an earlier part of a day (and it could be a brunch equivalent now).
    The Russian word for breakfast, zavtrak <= utro “morning” remained a morning meal … I guess it’s easier for morning meals to resist the upper-class schedule-drift pressures? Although if you let your day schedule shift *really* late, then breakfast is also trapped in the shifts. Like at dance festivals we attend, there is a perennial pressure to have breakfast served *early*, like at 4 or 5 in the morning, which is when people begin going to bed at last 🙂

  29. marie-lucie says:

    Tagalog as a curious example of meryenda shifting to an earlier part of a day
    Interesting. It could be that the word for a Spanish-style meal shifted to a different, more traditionally Philippine type of meal, served earlier.
    In one native Canadian community I am familiar with, the English word “lunch” does not mean only a midday meal but any type of informal meal, not one just cooked: for instance food carried in a “lunchbox” to eat on the job, a late night snack, etc. The crucial feature is not the timing of eating but the type of food and the informality.

  30. >Marie-lucie
    According to my Latin dictionary, “merienda” came from Latin “merenda”, a word that classical authors didn’t use. It’s related to Latin “mereo” (but finally from Greek “meiromai”: I obtain in the share-out), that means to win, to deserve. Our verb “merecer” is related to common Latin “merescere”, also from” mereo, ere”.
    The Latin verb “merendo” is to eat at midday or in the afternoon.
    On the other hand, our “meridiana” came from “meridianus”, related to “meridies” [medius, dies].

  31. BTW the 2005 Pomor dictinary is in the news again, since it won a foreign grant of $4,000, which potentially qualifies its creation, under the current Russian law, as an act of treason.
    The author, Igor Moiseev, chairs the Pomor Institute of Indigenous Northern Peoples of the Artic University. He used to espouse Pomor separatist views, although he must have mellowed more recently. Moiseev is presently being sued for extremism / fomentic ethnic strife but this has ostensibly nothing to do with his old publications. Instead, he’s alleged to have called the local non-Pomor Russians “быдло” in an anonymous blog comment traced to his computer.
    An investigation of treason (which would be a more harsh charge) might be motivated by the authorities’ need to justify eavesdropping of Moiseev’s phone.

  32. J.W. Brewer says:

    There can be drifts the other way, i.e. “noon” comes from “none” (= the 9th hour of the day = approx 3 pm). I believe the sorta-cognate (loan from Latin in both instances) word in Icelandic still means mid-afternoon. I assume one source of drift here is efficiency-minded medieval monasteries combining several of the appointed daily services into one and knocking them out earlier than their officially-appointed times of day so that they could get back to doing whatever it was they placed a higher priority on than worship.

  33. marie-lucie says:

    Gracias Jesús! I withdraw my comment, which was a hypothesis.
    If merienda has to do with “win, deserve”, perhaps it referred to a midday meal or a pause for eating given to the workers on an estate, in a workshop, etc. After a morning working hard, they would “deserve a break” and some food before resuming their work. Food would only be given to the “deserving”, so had to be “merited”.

  34. Nones is indeed the ninth hour (that is, halfway from mid-day to sunset) taken as a moment in time. But taken as a period of time, it extends from the sixth hour (mid-day) to the ninth hour; it was the former that got the name of noon.

  35. >Marie-lucie
    I agree with you as far “deserve”. On this matter I tell you an anecdote: a farmer and employer I knew said, speaking of his pickers of olives: “they even eat ham when they have a “merienda”!” However, the ham has been bought by them.
    About the time of “merienda”: building workers, country people, etc. at 10 or 11 (someone even says “tomar las 11”); children at 18 or 19.
    There is other nice word: “almuerzo”. It came from Arabic “al” and Latin “morsus”: the bite. Also it can mean a meal in the morning (breakfast or “merienda”) or a meal at 14 or 15 h.

  36. marie-lucie says:

    Some English people have a snack (?) called “elevenses” around 11, but I don’t know what they have at that time.
    In the area where I grew up, rural people, especially the men, would get up very early (before 6 am) and just have a cup of coffee and perhaps a little bread and butter, go tend the animals and work outside until around 10, then come back to the house where the women would prepare a fast-cooked meal more similar to a traditional English breakfast than a French one (with eggs, cold cuts, etc). They called this meal une collation, a word which can otherwise be used of a substantial snack. In some regions this is the word used of the snack given to children when they come home from school, but the more general word for what children have at that time is le goûter.

  37. It’s similar in rural Northern Germany where I grew up – farmers would get up very early (5 or 6 AM), and between breakfast and lunch they’d have a meal consisting of tea, bread, butter, cold cuts etc., between 9 and 10 AM. The term I know for this is zweites Frühstück “second breakfast”.

  38. Spanish (like other languages) has a similar word: “colación”, but hardly anyone uses it because the meanings related are:
    – light meal eaten at night to break one’s fast,
    -sweet things or cold meals to present a guest.
    Also we have “refacción” and “refección”: refection, snack, and “tentempié” ( as French “en-cas”; more o less “tenir debout”). There is only a short step from it to “tapa”.

  39. Trond Engen says:

    More or less the same thing all over rural Europe, I suppose. At least it was in the Shire.

  40. Collation in English means a light meal, often served to guests: it can be hot or cold. The word is not used much any more, at least in the U.S.
    Elevenses is a set of refreshments with tea or coffee.

  41. J.W. Brewer says:

    I became aware of “elevenses” at the age of say 9 or 10 because Paddington was always having elevenses with Mr. Gruber. I’ve never come across the noun in that sense elsewhere, so I infer from context that you can’t call a meal/snack “elevenses” unless a bear is somehow involved.

  42. marie-lucie says:

    For fans of (mostly) English meals through the ages:
    (I don’t know how to link directly).

  43. Thanks, m-l. here is the bbc link, and from there I discovered the concept of “National Wheatmeal Bread” (which apparently lasted in the UK from May 24th 1942 and into … 1956!). The Natl Bread was sort of a first, and flawed, incarnation of today’s popular wholegrain / multigrain breads (even though the name was “wheatmeal”, the British rationing authorities quickly moved to allow added barley and oat). I could only compare with the famed verse of Olga Berggoltz about rationed bread – the “125 Blockade grams” – which, as she puts it, consisted from as much blood and fire as flour.
    Another surprising discovery for me was that bleach in the low-gluten flour was not only supposed to make the dough whiter, but also firmer (by artificially oxidizing gluten and making chemical disulfide bond cross-links between gluten molecules). Despite my stint in protein chemistry, I never heard about this incredibly yucky-sounding cheat before. Potassium bromate, a class 2B carcinogen, is still allowed as gluten-oxidizer in the US (although supposedly in smaller concentrations to make sure that it decays more or less completely during baking). Life to learn!

  44. J.W. Brewer says:

    The BBC link is hyping an apparently new series featuring Clarissa Dickson Wright, who is the surviving featured member of the excellent Two Fat Ladies cookery tv series (which alas ended with the death of the other Fat Lady, Jennifer Paterson, in 1999).

  45. That leads to my dad’s claim that when he was in Japan in 1940, he passed the city (town?) of Usa, whose utility was that things manufactured there could be stamped MADE IN USA for export.
    My dad was a raconteur, so I’ve never been sure how much to credit this story.

  46. Thanks. Well, at least there is such a place, which means my father isn’t discredited totally: he never claimed it had been renamed. So unlike David Crystal’s claim to be addressed with singular y’all when he entered a Texas store as the customer fifty years ago, frinstance.

  47. Ooops. For “to be addressed” read “to have been addressed”, of course.

Speak Your Mind