When dialects started deleting unstressed and between two consonants ( such as when Myzithras became Mystras ), new clusters were formed and similarly assimilated by sandhi; on the other hand it is arguable that the dissimilation of voiceless obstruents occurred before the loss of close vowels, as the clusters resulting from this development do not necessarily undergo the change to [ fricative + stop ], e . g . ? ( ?? ) ?? as [ kti ] not [ xti ].
22.
However, the consensus among contemporary historical linguists is that the former explanation ( underlying representation ) is the correct one, in that an assumption of ATB would require multiple root shapes for the same basic root in different IE languages whenever an aspirate follows in the next syllable e . g . " d " for Sanskrit, " t " for Greek, " dh " for Proto-Germanic and Proto-Italic ( which have no dissimilation ) whereas the theory with an underlying aspirate allows for a single root shape with " dh " for all languages.
