I am looking for the exact history of the Spanish word esposas ("handcuffs") and its connection with the Latin word spondeo ("promise").

I read several times on the web the following story[a] (by example here) (my translation:)

When a merchant from ancient Greece made an agreement with a supplier, he sealed the contract by pouring a few drops of wine on the altar of one of his gods. The Greek word for this gesture was "spendo" (spilling a drink), but due to the habit imposed by the merchants, "spendo" gradually acquired the additional meaning of 'making an agreement' or 'signing a contract'. From "spendo", the Latin word "sponsus" was formed, used to name the person who assumes some commitment, just as today the person who commits to sponsor some initiative is often designated with the word taken from the English "sponsor" (perhaps more than with the Spanish sponsor). And if a man who promises to marry someone is a "sponsor", the woman who does the same is a "sponsa", a word that came into our language as "wife". The name "wife" given to the handles with which someone's wrists are imprisoned is a metaphor dating from the Middle Ages, by which the ideas of marriage and lack of freedom are linked.

In this text I don't understand why the Latin word spōnsus is explained from the Greek words σπονδή/σπένδω. Is the Latin root present in the words spondeo (and the late sponsāre) not enough to explain the words spōnsus/spōnsa ("husband"/"wife") ?

On the other hand, the link made in the Middle Ages between the meaning of promised and the meaning of handcuffs seems possible: can anyone give me precise references to medieval Spanish texts describing this shift in meaning?

[a] Cuando un comerciante de la antigua Grecia hacía un acuerdo con algún proveedor, sellaba el contrato vertiendo unas gotas de vino en el altar de alguno de sus dioses. La palabra griega para ese gesto era "spendo" (derramar una bebida), pero debido al hábito impuesto por los comerciantes, "spendo" fue adquiriendo poco a poco el sentido adicional de 'hacer un acuerdo' o 'firmar un contrato'. A partir de "spendo", se formó en latín el vocablo "sponsus", usado para nombrar a la persona que asume algún compromiso, así como hoy el que se compromete a patrocinar alguna iniciativa es designado, frecuentemente, con la palabra tomada del inglés "espónsor" (tal vez más que con la española patrocinador). Y si un hombre que se compromete a casarse con alguien es un "sponsus", la mujer que hace lo mismo es una "sponsa", palabra que llegó a nuestra lengua como "esposa". El nombre de "esposa" que se da a las manillas con que se aprisionan las muñecas de alguien es una metáfora que data de la Edad Media, por la cual se vinculan las ideas de matrimonio y de falta de libertad.

  • 3
    Yes, it is absolutely enough. That story is complete hogwash. Spōnsus is completely regular from spondeō, both the form and the meaning, and while the Latin verb is cognate with σπένδω, there’s no reason to imagine Greek merchants pouring wine on anything as part of the etymology of ‘spouse’. Commented Oct 14, 2020 at 20:48

1 Answer 1


To answer your first question, the page you cite is confusing cognacy with derivation. Latin spondeō and Greek σπένδω come from the same Proto-Indo-European root, but I see no particular reason to think the Latin word is a borrowing of the Greek word.

That said, Greek does shed light on the semantic shift, since it preserves both the original meaning "pour a libation" and the newer meaning "make a pledge, treaty, etc." (sanctified with a libation). Latin has lost the earlier meaning, but the semantic development is presumably the same.

  • 1. Is there a PIE-internal derivation to support the assumption that the earlier sense was "pour" (of libation)? I didn't think so. I find that to bind + s- rather implies bonding as the common root, but as uncertain as s-mobile is, it doesn't hold much conviction either. But the parallel from *spend- "libate" to *(s)pend- "stretch" is certainly notable, for it derives to spend [en.wt], shows s-mobile and has a tentative internal derivation in *(s)penH- "to turn" (note that generalia like turn, bend, cut, grow are nearly insignificant). In short, you are presenting with a...
    – vectory
    Commented Oct 16, 2020 at 14:08
  • ... certainty that you simply do not have. Or is there concrete evidence who poured what and when in prehistory? Not to mention that spray, sprinkle, sparkle, Ger. spritzen imply *spr- to me (tho fullgrade *sper- can be reconstructed; I am ultimately not sure, JLawler would simply speak of sound-symbolism, further see spunk, sputter, splatter, spit,...); syllabic r is known to tend to high vowels at that; further cp. speak ~ Ger. sprechen, also versprechen "pro-mis", the latter two probably from *sper-... For "handcuffs" anyway see Ger. Spange "clamp", reasonably from spann.
    – vectory
    Commented Oct 16, 2020 at 14:34
  • 1
    @vectory I don't understand what you mean by "a PIE-internal derivation". As I said, the sense "pour" is attested in Greek (and Hittite), as is the semantic shift "pour a libation" -> "pledge".
    – TKR
    Commented Oct 16, 2020 at 18:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.