/skandal(l)issa-/, sà-ka-ta-li-sà-

The imperfective suffix /-ssa-/ is deverbal, whence a verbal base *skandal(l)i- must be reconstructed. This non-attested base is a denominative either in /-i(ya)-/ (< *-i̯e/o-) or in /-(a)i-^(di)/ (< *-ei̯e/o-) from a non-attested adjective *skand-al(i)- (*-ó-lo-) or *skand-alla/i- (*-é-lo-).

[D.S.]

For the base lemma see Proto-Luwian *skanda-.

On the surface, Luw. *skandala/i- ‘gleaming application’, Gk. σκάνδαλον ‘trap, snare’, and Lat. scandula ‘shingle’ could be cognates. However, the phonological, morphological (cf. below) and semantic connection is apparently difficult.
If Gk. σκάνδαλον ‘trap, snare’ and Lat. scandula ‘shingle’ belong somehow to Luw. *skandala/i- ‘(gleaming) application’, the semantic development from ‘to gleam’ to the respective meanings must be made plausible.
Lat. scandula ‘shingle = application used to furnish the roof’ is semantically derivable from ‘(gleaming) application’ (cf. Luw. *skandala/i-) without difficulties.In N.H.XVI.15, Pliny the Elder refers to Cornelius Nepos who reports that until the war with Pyrrhus (280–275 BCE) Rome was roofed solely with shingles. For the types of trees used for producing shingles (glandiferous trees, esp. english oak [robur]; oriental beech [fagus]; resinous trees, esp. Scots pine [pinus]; white pine [picea]), cf. Plin.N.H.XVI.15, 18. Note that these trees provide relatively bright (= ‘gleaming’) wood.
Gk. σκάνδαλον ‘trap, snare’, if also traced back to ‘(gleaming) application’,Another possibility would be the root PIE 1.*skend- ‘to leap, jump (forth)’ (LIV²:554) by which Gk. σκάνδαλον ‘trap, snare’ could evolve via ‘stick in a trap which leaps when touched’ vel sim. (Gk. σκάνδάληθρον).could have denoted either an ‘application used for concealing a trap, lid’ or ‘toils, (ModHG) Lappen’.The latter denotes tools employed in the context of ‘enclosed hunting, (ModHG) eingestelltes Jagen’. This kind of hunting involves binding pieces of cloth or bunches of feathers (ModHG Lappen) to trees in order to drive the game into a given direction. The variant of ‘enclosed hunting’ confined to the use of (white and red) feathers (ModHG Federlappen) was known at least to the Romans, cf. Orth 1914a:570f. For hunting methods involving traps and snares in ancient Greece, cf. Orth 1914a:571f.

Gk. σκάνδαλον ‘trap, snare’ and Lat. scandula ‘shingle’, thus, can be interpreted either as inherited formations or as technical terms borrowed from Luw. *skandala/i- not later than the first half of the 1^st millennium BCE.

The phonological comparison of Gk. σκάνδαλον ‘trap’ and Lat. scandula ‘shingle’The vowel ‑u- of Lat. scandula results from weakening of any vowel in front of l pinguis, [ł], cf. Weiss 2011:117. The application of vowel weakening which proves the existance of *skandVla, a word for ‘shingle’ in Early Latin corresponds to the common use of this type of roofing in pre-republican Rome (cf. the comment above on Pliny).provides a potential common proto-form *skandalo/a-, which resembles Luw. *skandala/i- ‘(gleaming) application’ exactly.
Lat. scandula ‘shingle’ could be inherited from a form PIE *skend-eleh₂- and, therefore, represent an almost cognate of Luw. *skandala/i-.On the Latin root vowel -a-, cf. *skanda-, Morphological Reconstruction). However, as Schrijver 1991a:435 remarks, this change likely occured in Proto-Italic, viz. before the post-Proto-Italic sound change *‑n̥- > Lat. *‑en-, cf. Meiser 1986a:69f. This means that Lat. scandula cannot go back to PIE *skn̥d‑e‑leh₂-, but only to PIE *skend‑e‑leh₂-.If it is regarded as an inner-Latin formation, it is analyzed as scand‑ula-, an adjective with participial meaning.Cf. Leumann 1977a:311f. §283.1. The alternative interpretation as a deminutive (cf. Leumann 1977a:305 §282.A) is unlikely here.Finally, also borrowing before the time of vowel weakening is possible.
Gk. σκάνδαλον either is an inner-Greek innovationHowever, an analysis as σκάνδ-αλο- (on this suffix in Greek, cf. Chantraine 1933a:244–248) is very difficult to establish, because esp. the root vowel *-a- beside *-n- cannot be explained by existing sound laws. Furthermore, neither the root PIE 1.*skend- ‘to leap, jump (forth)’ (LIV²:554) nor the root PIE 2.*(s)kend- ‘(to begin) to gleam’ (LIV²:554) is attested in Greek with certainty, cf. merely the Hesychian gloss Gk. κάνδαρος ‘(Gk.) ἄνθραξ, coal’ (cf. [link skanda-]). PIE ?*sk̑end- ‘to appear’ (LIV²:546) is hard to connect for semantic reasons.or a loanword from Luw. *skandala/i-, because phonologically it cannot be derived directly from PIE.PIE *skn̥d‑e‑lo/eh₂- which is the assumed pre-form of Luw. *skandala/i- would result in Gk. †σκαδελο/η- vel sim.

The interpretation of Gk. σκάνδαλον ‘trap’ and Lat. scandula ‘shingle’ as loanwords on the one hand solves phonological and morphological issues, but, on the other hand, has disadvantages on the semantic side. Therefore, the question whether these two words are borrowed from Luwian and whether they somehow belong together is left open for discussion.

[A.O.]

(?)Gk. σκάνδαλον ‘trap, snare; (metaph.) stumbling-block, offence, scandal’ (Lxx Jo.23.13+),Cf. also Gk. σκάνδαλος ‘(Gk.) ἐμπόδισμος, hindering, impending’ (Ep.Rom.14.13, Hsch. σ 842), LSJ:547. gen. pl. σκανδάνων (PCair.Zen.608.7, 3^rd c. BCE)
(?)Gk. σκανδάληθρον ‘(Gk.) τὸ ἐν ταῖς μυάγραις, what is in the mouse-traps (Hsch.); stick in a trap (on which bait is placed and which shuts the mouse-trap)The stick in larger kinds of mouse-traps is called ῥόπτρον, cf. Poll.7.114, 10.156. This means that the traps provided with σκανδάληθρα are of a smaller kind.(Poll., EM); (metaph.) (word‑)trap (Ar., Cratin.)’ (Cratin.457, Ar.Ach.687, Poll.7.114, 10.156, Hsch. σ 841, EM.715K.44)Gk. σκανδαλάριος ‘shingler (of roofs)’ (IGRom.4.1646 [Philadelphia], Rhetor. in Cat.Cod.Astr.8(4).215, cf. LSJ:1604) is borrowed from Lat. scandularius ‘shingler’.
The inscription FD III.1.226 (2^nd/3^rd c. CE) attests the form σκανδαλιστης within an enumeration of terms for specific acrobats. Hence, this lexeme could have denoted an acrobat, too. Because Gk. σκανδάληθρον could designate a kind of stick, the interpretation of Gk. σκανδαλίστης as “trapeze acrobat” (cf. LSJ:1604, EDG:1341) might be obvious. However, since σκανδάληθρα were probably very small (cf. above), σκανδαλίστης is most likely derived from σκανδαλίζω ‘to cause to stumble’. Furthermore, the inscription is relatively late (2^nd/3^rd c. CE) and, hence, the evidence for the interpretation given above is fairly meager.
(?)Gk. σκανδαλίζω ‘to cause to stumble, cause to sin, give offence/scandal to, shock; (pass.) to be made to stumble, take offence’ (Lxx Si.9.5+)On this lexeme in the Septuagint, cf. Wagner 1999:289f.
GEW II:717f., DELG: 1010, EDG:1341f., LSJ:1604

(?)Lat. scandula ‘a wooden slat used for roofings etc., shingle’ (Hirt.Gal.8.42.1+)The affiliation with Gk. σκίδνημι ‘to scatter’ (PIE *(s)kedh₂- ‘to scatter, split’, LIV²:550f.), as proposed by IEW:919, is phonologically impossible, cf. also de Vaan (EDLIL:544) indentifying Lat. scandula ‘shingle’ with Lat. scindula ‘id.’ which itself belongs to scindō ‘to split, cleave’. However, because there is no model for scandula being a result of analogy, it is more promising to assume that Lat. scindula is a transformation of scandula under the influence of scindere ‘to split’. Derivation directly from Lat. scindere ‘to split’ is unlikely because of the semantics of the suffix (not †‘splitting’, cf. below). Lat. scandula ‘shingle’, therefore, has to be separated from scindere ‘to split’.
WH II:488f., EM:599, EDLIL:544 s.v. scindō, OLD:1699, Rich 1862:545