From Roger Lass’ Old English: A Historical Linguistic Companion, p. 133n.
Note how mere possession of a â€˜rich’ inflectional morphology doesn’t imply lack of ambiguity for any particular morphological class. The common argument connecting later loss of freedom in word order with erosion of noun inflection (order was stabilized to â€˜prevent confusion’) collapses here: except for some declensions in Gothic and Old Icelandic (Old Norse)), all Germanic noun plurals merge nom/acc, and they are rarely distinct in the singular. Yet the older dialects have a quite “free” word order.Clearly lack of morphological distinction even between such basic cases as those coding subject and direct object is not a â€˜problem’ for these languages, so the fixation of word order can’t be a â€˜solution’.
Pat’s note: I’m going to check out a book of his where he discusses this fully and will add to this entry from it. This concept he strikes down underlies most of the arguments of the grammar mavens for opposing any language change or even features that have been around forever but that they perceive to be changes.