Lat. quăssāre and its derivates
As we saw earlier, Lat. quăssāre, which meant ‘to shake violently, shatter, impair, disturb, etc.’, is a so-called ‘frequentative’ or ‘intensive’ verb derived from the basic verb quătĕre. As all Latin frequentative verbs, it is a first conjugation verb that was derived from the stem of the source verb’s passive participle, in this case quăss‑ (from quăss‑us) (cf. Part I, Chapter 8). This verb’s principal parts were present tense quăssō, present active infinitive quăssāre, perfect active quăssāvī, and passive participle quăssātus. In Medieval Latin, the verb quăssāre came to mean ‘to shatter’.
Lat. quăssāre was passed on patrimonially (by word-of-mouth) to some of the Romance languages. Thus, for example, it became quaissier and quesser in Old French, which has derived into different words, including Modern French casser [ka.ˈse], a very common verb now meaning ‘to break’, and Anglo-Norman quaisser (among other spellings), which was borrowed into English as quash, as we will see. Lat. quăssāre has been said to have given us two patrimonial Spanish verbs: quejar and cascar, each derived from a different Vulgar Latin version of Lat. quăssāre. Let us look at the Spanish and English reflexes of this Latin verb now in turn.
Sp. quejar [ke.ˈxaɾ] is now thought to have come from Vulgar Latin *quassiare, a modified version of classical Latin quăssāre, a variant that is not actually attested in writing, but which is presumed to have existed because of the form of this descendant in Spanish and Portuguese, as well as in other Romance languages, as we shall see. Actually, this is not the only theory as to the source of the verb quejar. The origin of the word quejar has been the source of much controversy over the years. As the great philologist Yakov Malkiel told us in the middle of the 20th century, ‘[t]he origin of Sp. quejarse (OSp. quexarse), Ptg. queixar-se ‘to complain’ and of their congeners has been one of the most widely discussed problems of Hispanic etymology’. We will proceed here with the assumption that quejar is indeed derived from Lat. quăssāre, as most scholars believe today.
In Modern Spanish, the verb quejar is strictly a pronominal (reflexively conjugated) verb, that is, it is always conjugated as quejarse (de), and its main meaning is ‘to complain (about)’, e.g. Los vecinos se quejaron del ruido ‘The neighbors complained about the noise’ or ¿De qué te quejas? ‘What are you complaining about?’. However, Sp. quejar can also mean ‘to moan, groan (from pain)’ in some contexts and thus can be translated into English by the verbs moan or groan, e.g. Se quejaba mientras le quitaban los puntos ‘He was making complaining noises as they were removing the stiches’ (Gran Diccionario de la Lengua Española Larousse).
The verb quejar appears very early in Old Spanish written records, such as the Cid, from the early 13th century and it was a rather common verb at the time. In Old Spanish, the verb was spelled quexar, pronounced [ke.ˈʃaɾ], or queixar. Note that the letter 〈x〉 spelled the sound [ʃ] in Old Spanish, the same as the sound spelled 〈sh〉 in English, as in the word sheep [ˈʃip] or cash [ˈkʰæʃ]. Around the 16th century, the sound [ʃ] in Spanish (spelled 〈x〉) changed its pronunciation to the sound [x] (‘jota’), which came to be spelled with the letter 〈j〉 in the 18th century across the board, except in words where this sound came from an earlier 〈g〉, in which case it is spelled with that sound (cf. Part I, Chapter 10).
In Old Spanish, quexar/queixar was primarily a transitive verb with a variety of meanings, primarily ‘to afflict, affect’, a sense that is today expressed mostly by the verb afligir or the derivate aquejar (see below). But there were other senses of quexar in Old Spanish, such as ‘to attack, pursue’, in a military context, ‘to urge, prompt, put pressure’, ‘to trouble, harass’, and ‘to prevail upon’, ‘to overcome’, etc. (see Malkiel 1945, p. 159). This multiplicity of meanings is one of the reasons scholars have been confused as to the origin of this word. The current theory, as we said, is that quexar comes from Lat. quăssāre (actually from a Vulgar Latin derivate, quassiare). The change in meaning from Lat. quăssāre ‘to shake violently, shatter, impair, disturb, etc.’ to the meaning quexar had Old Spanish, namely ‘to afflict, affect, etc.’ is not an unreasonable one. However, we can see why some might have come up with other options for the source of Old Spanish quexar.
As we said, quexar was typically transitive, as in Cuydados muchos me quexan ‘Many problems afflict me’ (J. Ruiz, 14th century, DCECH). It was occasionally used intransitively and then it meant something like ‘to be afflicted’ (Modern Spanish: estar aquejado) or ‘to lament’, e.g. Se lo ve quexar ‘One can see him (be) afflicted/lamenting’ (13th century, DCECH). In the 14th century, the verb quexar stopped being used transitively, however, and came to be used only pronominally (reflexively) and, thus, intransitively. Note that in those days, reflexive quexarse did not mean ‘to complain’ like quejarse does today but, rather, the it meant the same thing as intransitive quexar, namely ‘to be afflicted, upset, distressed’ (DCECH), ‘to show distress’ (Malkiel: ‘to cry, weep, wail, sob’). The meaning change of quexarse from ‘be distressed’ to ‘complain’ is a reasonable, unsurprising one. The new meaning can be found in writings as early as the beginning of the 14th century (DCECH) and it is the normal meaning of the verb by the 15th century.
Let us look now at the speculation that Old Spanish (O.Sp.) quexar is derived from a putative Vulgar Latin *quassiare and not from Lat. quăssāre. The reason is that we would have expected an Old Spanish word derived by word-of-mouth transmission from Lat. quăssāre to have resulted in *casar [ka.ˈsaɾ], or perhaps even *cuasar [ku̯a.ˈsaɾ], but not quexar [ke.ˈʃaɾ]. That is the reason that the unattested Vulgar Latin derived verb *quassiare has been postulated. The i in this word represents a semi-vowel [i̯] sound that would explain the change of the alveolar consonant [s] to the post-alveolar or pre-palatal [ʃ]. And this change itself explains the rise of the vowel [a] to [e], for the low vowel [a] was typically raised to [e] when next to a palatal consonant such as [ʃ]. The loss of the semivowel [u̯] after the consonant [k] is expected in Old Spanish patrimonial words between [k] and a high vowel [i] or [e] (cf. Sp. que [ke] and quien [ki̯en]), but also before an unstressed [a], e.g. Sp. calidad < Lat. qualitatem, so the e is not really needed to explain this loss. (For all these sound changes, see Part I, Chapter 10.)
Interestingly, the existence of an unattested variant *quassiare of Lat. quăssāre in Late Latin or Vulgar Latin is corroborated by a word in another Romance variety, namely Old French, whereas a cognate of this Spanish verb is attested as quaissier, quesser and quasier in different dialects. In some dialects of Old French, the cognate of this verb was quasser, a verb we mentioned earlier as the source of Modern French casser ‘to break’. The variant with the i in it, which caused the consonant to be palatalized, was the one used in Anglo-Norman, the variety of Old French brought to England, which resulted in the English loanword quash, that we shall discuss below.
 Quoted from p. 142 of “The Etymology of Hispanic que(i)xar”, by Yakov Malkiel, Language, Vol. 21, No. 3 (Jul.-Sep., 1945), pp. 142-183, Published by: Linguistic Society of America.
Malkiel himself argued that O.Sp. que(i)xar, which according to him originally meant ‘to oppress, to squeeze’, is derived from the O.Sp. noun qu(i)xo ‘jaw’, which is not a popular theory today and DCECH does not even mention it. Among the other theories of the origin of O.Sp. quexar, there is one that says that it comes from a Vulgar *questare, a frequentative of querī ‘to complain’, but that has been dismissed as a phonetically impossible derivation (DCECH). For more early theories on the origin of O.Sp. quexar, see Malkiel’s 1945 paper or the entry quejar at the DCECH.
The noun queixo ‘jaw’ that Malkiel postulated as the source of quexar is patrimonially descended from Vulgar Latin capseum ‘box-like’, from capsa ‘box’ (source of Sp. caja ‘box’), cf. Sp. quijada ‘jaw, jawbone (of an animal)’, from Old Spanish quexada. From this noun queixo, ‘warriors of Christian Spain’ presumably derived the verb que(i)xar ‘to press between the jaws’ and subsequently ‘to squeeze, to crush, to smash'’ between the 8th and the 10th century (op. cit. p. 180).
 As Malkiel tells us, ‘The ratio of frequency between quexar and quexarse in narrative texts of the 13th century is 10:11; in the early 14th century book Confisión del Amante (c. 1400) it changes to 0:18’ (op. cit., p. 158).
 Etymologists explain the palatal sound [ʃ] at the end of Eng. quash as being caused by the front (‘palatal’) vowel i in the French word, which was not there in Latin. The OED explains things the following way: ‘The usual modern form with final /ʃ/ reflects French forms with a palatalized sibilant (represented in spelling by ‑iss‑ or ‑ssi‑ ); such forms are especially frequent in Anglo-Norman, while in continental French they occur only in the far north (Picardy, Walloon) and east (Lorraine, Moselle, Switzerland). An unattested post-classical Latin etymon *quassiare ‘to crush, break, shatter’, variant (with insertion of vocalic glide) of classical Latin quassāre, has been posited to account for such forms in French (see Französisches etymol. Wörterbuch s.v. *quassiare), but it is possible that they arose by group analogy with other verbs also having palatalized forms’ (OED).