*h₁elmos: "wych elm (Ulmus glabra)" shows up with identical meaning in all four of the Late Northwest branches, plus an Albanian with a slightly different meaning. It probably represents a LNW borrowing from an Eastern European substrate language, with Albanian borrowing the term separately from either the same or …
A quick Austronesian word family inspired by my friend Kai's birthday: Malayo-Polynesian *tasik: "sea, coastal water, saltwater, tide" -> Hawaiʻian kai: "sea, saltwater, coast, tide".
The Oceanic compound *taqu-tasi: "expert sailor" (lit. "sea-able, sea-knowing") is such a cool word, I included Austronesian *tsaqu: "to know how, to be able", too …
Étienne de Silhouette, a French politician of Basque ancestry, was the Controller-General of Finances under Louis XV in 1759. His measures to curb the national debt, notably by imposing new taxes on the rich and the nobility prompted a backlash and ultimately his dismissal after less than a year …
There are two words from Sumerian whose descendants are still used all over the world today: gi: "reed"—what they wrote with—and dub: "tablet"—what they wrote on.
Just in case you wonder how important Sumerian writing was in world history.
Proto-Indo-European *h₂ew: "away from, since, again" cannot be formally tied to *h₂epó: "off, away", but the similarity is suggestive. Perhaps Pre-Proto-Indo-European *b became either *p (by fortition) or *w (by lenition) in almost all cases, accounting for the paucity of *b in PIE.
Dr. Melenaite Taumoefolau makes a compelling and intriguing argument: that Nuclear Polynesian *sawaiki: "traditional/mythological homeland"—whence Hawaii—may be a cognate of *sau ʔariki borrowed from the Tongic branch into Nuclear Polynesian. (Note that *sawaiki can only be reconstructed to NP, it is not present in Tongic or …
The etymology of dragon (Classical Greek δράκωνdrákōn) is not certain, but is traditionally linked to δέρκομαιdérkomai: "to see, to watch".
"Derkomai…really?" Not so odd: The aorist active participle (m.) of δέρκομαιdérkomai is δρακώνdrakṓn—the only difference from δράκωνdrákōn is the placement …
I've got three families, each a possible origin for West Germanic *panning: "penny", possible borrowings from either Latin patina: "pan" (via variants *patna, *panna), Latin pannus: "cloth, or—less realistically—Punic 𐤐𐤍pene: "face".
I prefer the explanation that the first pennies were small …
The attempt to brand cis(gender) as a slur is not truly about the word itself—they wouldn't be any happier with any other word that was used for the same meaning—but an attempt to deny the very category of "cisgender" and thus by extension the existence of …
There are two west Asian wander words for musical instruments: *pandVr- and tabVr-. *pandVr- mostly means a stringed instrument, and tabVr- mostly means a kind of drum, but that is not consistent.
The *p-d and *t-b could possibly be metathesis of place of articulation, while leaving voicing in place …
I've included (possibly) two families again in this one, Proto-Indo-European *wet-: "year, year-old" and Proto-Indo-European *wósr̥: "spring (season)". There's no formal connection, but it's not hard to imagine on either the phonetic or the semantic level.
Acrostatic *ó-r̥ nouns like *wósr̥ are from very early PIE. You could imagine …
This word for "moon" is wide spread through Afro-Asiatic and specifically Semitic languages. The one exception is Arabic which uses قمرqamar for moon, instead of the expected *war(i)ḵ. The origin of qamar is unsolved.
Though Arabic may have borrowed the word tārīk: "date, time, time-keeping" from …
In which linguistics should probably be *dinguistics.
Armenian լեզուlezu, Lithuanian liežùvis, and Latin lingua (all "tongue") are modified from their expected reflexes of *dn̥ǵʰwéh₂s by association with *leyǵʰ-: "to lick". The expected words would be Armenian *tezu, Lithuanian *iežùvis, and Latin dingua (attested in Old Latin).
Both Classical Sanskrit 𑀫𑀭𑀓𑀢marakata: "emerald" and Classical Greek μᾰ́ραγδοςmáragdos: "emerald" appear to be borrowed from a source like *maraktV.
This is presumably a form closely related to the Phoenician brqt and Hebrew baréket: "emerald", from *b-r-ḳ: "shining, sparkling". Akkadian cognate baraqu: "shine" instead might have had …
The American Indigenous languages from the US and Canada are mostly extinct, dying, or spoken by relatively tiny remaining communities. It is easy to assume the same about Central and South America, and to think of some of the well-known historical peoples—like "Incas", "Aztecs …
It is very tempting to unify *pəñu: "sea turtle" with *pəNuʔ: "to fill, full, container". And *pəNuʔ: "full" has potential cognates in the possible Sino-Austronesian macrofamily/Southeast Asian historic sprachbund.
And then at that point you're speculating a link between "green sea turtles" and …
Turtledoves are not named after turtles, in fact "turtle" has meant the bird for longer than it meant the reptile.
In Middle English tortu meant "turtle" (from Latin tortuca) and turtel mean "turtledove" (from Latin turtur), then they collapsed into "turtle". The modern meaning of "turtle" won out, and …
A Mediterranean/Caucasian substrate word approximately like *wion: "violet (flower)", was borrowed into Hellenic, Iranian, and Italic.
Greek Ionia is probably not related, since the Mycenaean shows 𐀂𐀊𐀺𐀚 i-ja-wo-ne for "Ionians" with digamma in a different position than the *wion expected for "violet" , and Egyptian 𓇌𓅱𓈖𓏭𓉻𓂝𓏛ywnj-ꜥꜣ supports that …
Here I'm following Blažek's compelling argument (2017) to derive *h₂ŕ̥Tḱos from *h₂r̥dh₂éḱh₃-s: "bee-eater", similar to many later words for bear that mean things like "honey eater", "honey pig", "thief of bees", "bee bear", "bee wolf", … . This instead of the traditional connection to Sanskrit 𑀭𑀓𑁆𑀱𑀲𑁆rákṣas and …
The PIE word *yóh₁r̥: "year" becomes both the native English "year" and also the Latinate "hour". Fascinatingly, Greek ὥρᾱhṓrā (whence the Latin) is also borrowed, post-Alexander, into Classical Sanskrit along with the Hellenistic-Babylonian astronomy/astrology, eventually giving such descendants as Thai โหรhǒon: "astrologer, prophet, magician"!
I've been thinking about herons and egrets a lot recently. For the past year or so, our morning walk goes by a half mile of drainage ditch with an amazing variety of water birds. We regularly see 4 different species of heron: great blue heron, great egret, snowy egret …
The period of unseasonably hot weather after an initial autumn cooling—now often called "Indian summer"—was previously called "goose summer" in English because it corresponds with the beginning of goose hunting season. Compare Irish fómhar beag na ngéanna: "warm period in Autumn (Indian summer)", literally "little autumn of …
While *linom: "flax" and *līnom: "flax" obviously must be connected in some way, the difference in vowel length cannot be reconciled by any regular changes.
All versions with the long vowel are plausibly attributable to derivation from Latin līnum. All the short vowel versions could theoretically be from Greek …
*ḱel-: "to lean", which gives us *ḱlew-: "to hear". The connection is most explicit in *h₂eus-ḱl̥teh₂ye-ti, literally: "leans an ear", which gives Latin auscultō: "I listen, I hear, I heed, I obey".
The colors white, gold, yellow, yellow-green, green, sky blue, and navy blue in various languages. And the chemical elements chlorine, gold and arsenic in English.
Proper Freedom for our Friends this Friday (Juneteenth). ✊🏿✊🏽✊🏼
And perhaps more importantly, proper freedom even for those who are not our friends. Which is in contrast to the ancient, but all too modern, semantic connection where only those we like get to be free.
This is quite a large family with a number of borrowings into other language families. Because, it turns out, the Indo-European speakers' relationship with wheels is pretty important in a lot of people's history, like the Sumerians' relationship with reeds.
*kʷetwer- is too long for a root. The feminine root suggests that there may have been an original root of just *kʷet-. There are also several reasons to believe that none of this is the original word for four:
Hittite's word for "four", meyawes, is unrelated (Hittite …
Celtic, Greek, and Latin each have words for "island" that involve /n/ and /s/ and cannot be completely explained. One possibility is that they are all borrowings from the same non-Indo-European source.
Even if those three are all connected, it would still be beyond credibility for Malayo-Polynesian *nusa …
"Warlock" originally meant "oathbreaker", a compound of *wērō: "truth, oath" and *lugô: "liar". Here I have the families for both pieces of the compound.
There also used to be a related word in Old English þēodloga, which I am quite fond of. It would have been thedelock in English …
It has long been noted that the Ancient Greek λέωνléon: "lion" (whence the word for lion in nearly all Balto-Slavic, Germanic, and Italic languages, as well as a few others) looks almost like it comes from Semitic *labu, but not quite. Pre-Classical Greek would be …
There's a very cool fact: cabbage, broccoli, cauliflower, kale, Brussels sprouts, collard greens, kohlrabi, and others are all the species of plant, namely Brassica oleracea, selectively bred for different parts of the plant. (https://www.google.com/search?q=brassica+oleracea+selection)
One day, I picked up my phone to look up the etymology of Pontic, but got distracted and looked up the etymology of path first. And it turned out they were the same!
But wait, how does that work, cause Grimm's Law? Germanic *paþaz is apparently a borrowing from …
Indo-European seems to have had two words for "blood" *krewh₂-: "spilled blood, blood outside the body" (with extended meanings of bloodshed and cruelty) contrasts to *h₁ésh₂r̥: "flowing blood, living blood" (with extended meanings of family).
The root *krews- is probably a derived root, and means "to harden, to coagulate …
(Tegurala introduction: I have a D&D setting I have developed, which I call Tegurala. When I started working on it, I decided to use real world languages for the fantasy languages, for naming language consistency without having to create a bunch of new ConLangs. Giants use Uralic languages)
Proto-Indo-European has *gʷṓws: "cow, cattle" and *woḱéh₂: "(female) cow". The expected feminine of the *gʷṓws would be something like **gʷow(s)éh₂. *woḱéh₂ looks similar, but not quite there.
Sino-Tibetan *ŋwa: "cattle, ox" is also highly reminiscent of *gʷṓws. If there was a Central Asian …
I present two families for "pig" here, partly because Armenian խոզxoz is derivable from either family (see footnote [1])
Family 1, Indo-European *suH-pig, sow?", possibly related to Akkadian 𒊺𒄷𒌑še-hu-u₂ and Sumerian 𒋚šah; and family 2, Proto-Indo-European *ǵʰóryos: "pig", probably central PIE …
I get a little speculative here, if not downright provocative. There are words for dog all over Eurasia that recall each other. These could be onomatopoeic, but they do not much resemble the common forms you find in words that actually used as dog sound …
The Greek and Armenian words are often interpreted as being from a Mediterranean substrate word. But then there's the Slavic *lȏjь: "tallow, suet" and Old Chinese lɯw: "oil, grease", which is plausibly from the Tocharian equivalent of Slavic *lȏjь, and …
I'm in the process of expanding this family, after merging the original Proto-Indo-European *bʰleh₃-: "flower" with Proto-Indo-European *bʰelǵʰ-: "to swell" (originally under title word a href="/word-family-budget.html"Budget/a)—and putting both under Proto-Indo-European *bʰel-: "to blow, to swell up".
Pterodactyl. Two cool realizations about Greek "pter-". 1) "Helicopter" is so strongly segmented as "heli-copter" in English, it's fun to realize that the Greek construction is actually "helico-pter": "twisting (helix) wing". 2) More recently I realized it's an awesome example of Grimm's Law: apply …
Ankylosaurus! And as an unexpected bonus, I found Azhdarchid pterosaurs deep in the family, too. Pterosaurs are probably my very favorite Mesozoic animals (stay tuned for next week!). Also in this family: various spiders and snakes and fears and sins in various languages.
This is the 100th Word Family Friday in 100 weeks, and the last before I take a break for a few weeks to work on infrastructure and pad my buffer. I originally had a different idea, but then I realized I of course had …
*ksweyb- is a _very_ weird root. It has too many consonant, and way too many in the onset. It has a /*b/ which is an unusual/marginal phoneme in PIE. It is related to a number of …
Latin words munio: "fortifiation" (English munitions) and murus: "wall" (English mural) appear to come from a *mey- root, so could be related, but are more likely connected to *mewtʰi- which gives various words about "pole, stick, post, picket"
The Celtic word *dūnom: "hill fort" was borrowed into Germanic twice, once as "hill" and once as "fort".
The first time was borrowed before the completion of Grimm's Law, so Grimm's Law changed the *d to *t for Germanic *tūną: "enclosure", which eventually gave English "town …
This word for "guest"/"host" is restricted to the Late Northwest Indo-European languages, appearing—as far as I know—only in Balto-Slavic, Germanic, and Italic, plus a single personal name in Lepontic (Celtic). It seems to be cognate to a word for "to eat" which is …
"My language is my treasure and my prized ornament."
This page describes a counter-factual in which the Tengwar writing system (from Tolkien's Middle-Earth) is instead a native writing system—called Tuʻara—developed by speakers of Polynesian languages.
For the end of my architecture theme month, I've put together a collection a few small-but-interesting families with architecture words that I did preliminary research on, before concluding each was too small to a weekly family on its own.
Apologies for the sketchy nature. This one exploded on me in several different ways.
Prior to working on this family, I knew that "ambassador"/"embassy" was from Gaulish via Latin, and I had connected that to the story Livy tells about the Battle of …
June's theme is linguistic effects of the Celtic Hegemony of the early Iron Age. The rest of the month will focus on words that were borrowed from Gaulish into Latin (things like very iconic Latin word gladius being borrowed from Gaulish, which appears in …
Proto-Indo-European *h₂éwis: "bird" and Proto-Indo-European *h₂ówis: "sheep" appear to be umlaut variants, presumably through the more sparsely attested root *h₂ew-: "to put on clothes, to cover". *h₂éwis: "bird" would be from a passive participle, "a clothed (in feathers)". *h₂ówis: "sheep" from an active participle, "a clother (in wool)".
In English, constituent question words are often called wh-words: what, when, where, which, why, whether. All these are spelled with a w and pronounced with a w (the h is pronounced in some dialects and not others). This is not coincidence.
Germanic really went to town with this root, creating a bunch of morphological forms that aren't attested in any other branches, and in a few cases I can't even relate them to any standard derivational forms I know in either PIE or Proto-Germanic.
This is the family that no matter how long I work on it, it's never done. That's what I get for trying to explore the possibility that two extremely important and productive roots may be connected: *h₂er-, related to cosmic order, and *h₃reǵ-, related to societal …
*bʰewdʰ- vs. *gʷʰedʰ-: two families that end up being super confusing in Germanic languages, since *bʰ and *gʷʰ merge at the beginning of a word. In English, they end up being homophones with complementary/opposite meanings: "offer" vs. "ask for".
We start religion words with *dyḗwos ph₂tḗr, Old Father Shining-Sky himself.
Interestingly English "day" is not from this root, despite similarity to many IE words for day, e.g. Latin diēs, etc. Germanic d corresponds to Latin f (PIE dʰ); Latin d corresponds to Germanic t …
A root meaning "to run, to depart" can be reconstructed from Hellenic and Indo-Iranian. But in Northwest branches, Balto-Slavic, Celtic, Germanic, and possibly Italic, there is a phonologically identical root meaning "to die". So it's probably a euphemism coined in a Late Northwest Indo-European …
I'm not seriously proposing that words for "death" all across the Northern Hemisphere are related, but I'm not entirely not proposing that, or at least that there may be an ancient wander-word (possibly driven by taboo replacement?) that made its way across much Eurasia …
Sumerians and reeds were very influential! (Cuneiform was written with a sharpened reed stylus and was the first writing system, possibly tied with Egyptian hieroglyphs.)
*péh₂wr̥ means "fire" as an inanimate substance (grammatically neuter), as opposed to *h₁n̥gʷnis, which is animate, elemental fire (grammatically animate in earlier PIE, masculine in later PIE).
*péh₂wr̥ is though to contain the (é)-wr̥ suffix that derives nouns from roots, though there are no …
Proto-Indo-European had (at least) two words *h₁n̥gʷnis: "fire" (grammatically animate) and *péh₂wr̥ (grammatically inanimate).
Effectively all descendants of *h₁engʷ- have lost the labialization, so distinguishing it from reconstructed **h₁eng- is not immediately obivous. It mostly survives only in languages that merge *g and *gʷ (Balto-Slavic …
This one is really cool and just might go back all the way to the original domestication of horses in Central Asia. Maybe—just maybe—the Thai word for the knight piece in chess and the Welsh word for the knight piece in chess might be …
*wed-: "water" and some potentially related other roots
I love how Indo-European *udrós, "aquatic", became the word for "otter" in 4 out of the 5 branches where it is attested, except for Greek where it became "sea serpent".
Semtic *ʾartɬʼ-: "earth" and Proto-Indo-European *h₁erTeH: "earth". I connect these for several reasons (beyond the obvious). First, there are several indicators that *h₁ was pronounced as a glottal stop, including a number of other words that are apparent correspondences between PIE *h₁ and Semitic …
*ḱer(h₁)-: "to grow, to make grow, to increase" is mostly in the western IE languages; *ḱer-: "to braid, to weave, rope" is mostly in the east, with only Greek and Armenian having both forms. I would suspect the western *ḱerh₁-: "to make grow" form is a derived root …
All words mean "seven" unless otherwise specified.
It may seems strange to see the word for "seven" borrowed around between so many different language families, when the numbers 1-10 are so highly conserved in descendants of (post-Anatolian) Proto-Indo-European …
A small family, apparently from *dʰeh₂bʰ-: "fit together, fitting, make".
Germanic *(ge)dabaną: "to fit, to be fitting" and Balto-Slavic forms like Lithuanian dabinti: "to decorate, to beautify" and Slavic *dobrъ: "good" are almost certainly related to each other. Likewise Latin faber: "craftsman, maker, smith" and …