Biblical Hebrew Forum

["Vadim Cherny" <VadimCherny AT mail.ru>]

> > The "word" in the above rule probably means a prosodic word so that in A
> > maqaf B the first letter in B wouldn't have a dagesh.
>
> It's not a prosodic rule.  It's not hard to find examples in Tanach of
> the same prosody with different trope types, and, hence, different
> BEGED KEFET behavior.  The Masoretes invented this arbitrary
> inter-word rule (and, actually, for all we know, invented the entire
> BEGED KEFET rule).

The people who painstakingly marked every minutest thing in the Tanakh 
invented a rule, huh?
They just wrote down what they actually heard. And, of course, they heard 
strengthening (loss of aspiration) after a stop. And, of course, anyone 
concerned with distinct pronunciation in singing (cantillation) introduces 
stop between the words to avoid concatenation and subsequent loss of first 
consonant in a cluster.

natan bakar -> nata nbakar (nata 'bakar) -> natan.bakar

This is the same process as created intra-word dagesh kal:

ni-zcar - ni'car -> niz.car

This makes clear why maqef+vowel eliminates dagesh: consonant clusters 
appear, no blurring of the first consonant, nothing to prevent. Since the 
first word in maqef constructs loses accent, its final vowel is almost 
unaccented, and there is no post-tonic gemination (dagesh hazak), either.

Vadim Cherny