With regard to the ultimate etymology, a suggestion has been made by Prof. Skeat that Middle English gunne may represent a hypocoristic form of a Scandinavian female name compounded with Gunn-. This conjecture receives a strong confirmation from the fact (communicated to us by Mr. W. H. Stevenson) that an account of munitions at Windsor Castle in 1330-1 (Exchequer Accts. Q.R. Bundle 18, no. 34, Pub. Rec. Office) mentions ‘una magna balista de cornu quae vocatur Domina Gunilda’. There are other instances of the practice of bestowing female personal names on engines of war; but there was no distinguished lady named Gunilda (= Old Norse Gunnhild-r; spelt Gunnild in Havelok) in the 14th cent., and it seems highly probable that this use of the name may have come down from Scandinavian times, when its exceedingly appropriate etymology would be understood (both gunn-r and hild-r mean ‘war’). If Gunnhildr, as is likely, was a name frequently given to ballistæ and the like, it would naturally, on the introduction of gunpowder, be given also to cannon. Indeed, there is some appearance of evidence that an explosive engine was actually called by this name many years before the earliest recorded instance of the use of gunpowder in warfare. The ‘song against the retinues of the great people’ in Pol. Songs (Camden) 237, which must have been written in the reign of Edward II, contains the following passage
The gedelynges were gederedThe correct translation of this passage, which has hitherto been unexplained, seems to be as follows < ‘The lackeys were gathered out of Gunnild's spark [Old English gnást: see gnast n.]; the grooms and pages, the varlets with their boasting, all were hatched of a horse's dung’. According to analogy, the regular ‘pet-name’ in Old Norse for Gunnhild-r would be *Gunna, which would give Gunne in Middle English; Rietz Sv. dial.-lex., mentions Gunne as a female Christian name still surviving in Swedish country districts. (In Iceland Gunna is now common, but it is taken to stand for Guðrún.)
Of gonnylde gnoste;
Palefreiours ant pages,
Ant boyes with boste,
Alle weren y-haht
Of an horse þoste’.
The other suggestions that have been made as to the origin of the word are obviously unsatisfactory. The assumed Old French *mangonne, of which gonne has been supposed to be a shortening, is wrongly inferred < mangonneaumangonel n., and is not philologically possible, unless as a back-formation. The French gonne, large cask, does not occur before the 16th cent., and is regarded by Littré as adopted from the English gun. The conjecture that Middle English gunne is of echoic origin perhaps involves no impossibility, but it has no positive support, and little intrinsic probability.NB "thoste" is defined as "dung, excrement, a turd"; can't find "gedelynges" though (obv. no connection w/ Gatling!).
Update As Alan tweeted in response to this post, "Wat Skeat shoots again!"