I think it's funny that in French the word for "unicorn" is "licorne" because:
- The word "unicorne" was first reanalyzed as "une icorne"
- The definite article was then added, making it "l'icorne"
- The new definite form was reanalyzed once again, resulting in "une licorne"
Before any anglophones get on the French people's case on this, consider for a second what y'all did when you reanalyzed the Spanish "el lagarto" ("the lizard") as "alligator."
Reanalysis is fun.
Oh yeah, everybody does this*. Another English example is "apron", which was once "napron" until we reanalyzed the initial N as part of the indefinite article (a napron -> an apron).
A fun one in Arabic is the city of Alexandria in Egypt. Quite understandably, Arabic speakers heard the initial "Al" and thought "ah yes, the ubiquitous definite article" and Alexandria became al-’Iskandariyya.
In the opposite direction, Spanish adopted hundreds of Arabic words during the Middle Ages due to Andalusian/Islamic influence, and there are very few Spanish words that start with al- that aren't of Arabic origin (and in fact, many words that start with A without being followed by an L, as in about half of cases in Arabic the L in "al-" is elided).
Reanalysis occurs in many other places besides article-noun combos, of course, but it's an extremely common case.
*citation needed, but reanalysis is extremely common
Oh, this actually explains something I'd just attributed to a quirk of sequence constraints or something; why Alexander is realized as Iskander/Iskandar in Arabic! It makes sense to analyze it as al-Iskander in Arabic!
Same thing happened with the word alchemy! Started out as the Arab term "al-kimiya", and when it was transported to Europe, it became "alchemy". This is actually really interesting, because as the term evolved more, it became "chemistry", effectively un-reanalyzing the word!
Oh actually there's another layer of fun there: the Arabic "al-kimiya" is actually a loan of the Ancient Greek χῠμείᾱ (khumeíā), which was used to refer to the art of alloying metals. Arabic borrowed a lot of Greek terminology owing to Arabic translations of Greek classics (many of which were actually lost in Europe until they were retranslated from Arabic). So, yeah, the Greek khumeíā made a round trip through Arabic, then into medieval Latin as "alchemia," and from there we eventually do get chemistry!
Not quite the same thing, but this reminded me of one of the funniest phenomena in the German language.
So, you may or may not know that x-rays were discovered by a guy called Röntgen (or Roentgen, though the ö is the proper spelling). Because of that, they're called "Röntgen rays" in German. Now, the thing is that in the German, the infinitive of a verb is always formed with an -en at the, so, for example, "to run" is "rennen" or "to sleep" is "schlafen." And because of that, it just so happened that the verb for performing x-rays became... "röntgen."
ich röntge, du röntgst, er/sie/es röntgt, wir röntgen, ihr röntgt, sie röntgen
In the X-rayed lab, straight röntgin it
I love reanalysis, so very much.
Swahili has a series of prefixes which indicate noun class, agreement, various things. Basically, every noun gets a prefix, and there are many different sets; each class has a dedicated singular and plural form.
One of the many classes is ki-/vi-: kitanda (bed) vs. vitanda (beds), kiti (chair) vs. viti (chairs), etc.
Swahili has a number of loanwords from Arabic. One of these is the word for book - in Arabic, that's kitāb ﻛِﺘَﺎﺏ .
...which Swahili naturally looked at and said, "Oh, excellent, it's a ki/vi noun!" Thus the Swahili word kitabu (book), and its plural form, vitabu.
One interesting example I came across recently is in Kagayanen, a language spoken in the Phillipines. It has a ma- prefix that can derive an adjective from a word that wasn't one to begin with (asin "salt" -> masin "salty"), or alter the meaning of an existing adjective (lised "difficult" -> malised "sorrowful") The spanish adjective maligno "malicious" was borrowed into Kagayanen as maligna "defiled/disgusting", then the ma at the start of the word was reanalysed as being the adjectival ma- prefix, and the word ligna "filthy/evil thing" was born. Source: pg138