According to Blust (2007) and Smith (2015) voiced implosive reflexes of PKen *bh *dh, *jh, and *gh justify a Lowland Kenyah group of languages, as opposed to Highland Kenyah where the same PKen phonemes are reflected by voiceless stops.
In the following section I will discuss two specific statements regarding the shape of PKEN and offer evidence from Western Lowland suggesting that if anything, the conservative phonology of Proto-Kenyah has been slightly understated and some of the sound changes previously assigned to Proto-Kenyah are likely parallel innovations.
PKEN *dua *panah Pre-PPS *-h > 0 - *pana PPS *-a > *-ah *duah *panah Sebop duah **panah The Sebop word for 'hot' is pana, not **panah, which is only possible if PPS had retained word final *h.
Blust (2007:7) states that PNS *u and *i became o and e in final position in PKEN, where *h (from PNS *R and *s) had been lost.
Penan share distinct lexical and phonological innovations exclusively with each other, but among Western Lowland languages, Eastern Penan alone contains reflexes of PKEN *tuju[??] 'seven' and *pulu[??] 'ten.' Also, among Penan-Sebop languages, Eastern Penan alone contains reflexes of PKEN *luton 'fire', *bulu[??] 'bamboo', *iap 'chicken', *edhaw 'day', *endum 'nose', *madan 'fly', and *babuy 'wild boar.' It appears that the changes which define both Western Lowland and Penan-Sebop were underway, but not yet complete, when Penan split from Penan-Sebop.
Evidence from Lowland Kenyah languages suggest that PKEN had not deleted *h in final position, nor had it lowered high vowels before word final *h.
English awake bamboo banana PMP *banun *buluq *punti PNS *bulu[??] *putti PKEN *taga *bulu[??] *peti Sebop tu[??]ut lepek balak E.