A couple of things I noticed

[alice]

...the universal constraint that plurals may not show a greater degree of morphemic contrast than the corresponding singulars... (p105, #3)

Is this actually a thing, or does it not mean what I think it means, that morphology is always more complex in singulars than plurals? It's not hard to find counterexamples

And, a reputable source which I annoyingly can't remember says that Latin got articles from Greek, which got them from Egyptian. To what extent can this be considered true?

[Raphael]

And, a reputable source which I annoyingly can't remember says that Latin got articles from Greek, which got them from Egyptian. To what extent can this be considered true?

I don't know enough to tell if there's any truth in it, but it sounds to me like something someone with a very pop science understanding of languages (I wouldn't even say linguistics) might come up with.

[Lērisama]

And, a reputable source which I annoyingly can't remember says that Latin got articles from Greek, which got them from Egyptian. To what extent can this be considered true?

The extent to which correlation = causation?

Edit: removal of accidental interrobang¹
¹ I got carried away

[WeepingElf]

And, a reputable source which I annoyingly can't remember says that Latin got articles from Greek, which got them from Egyptian. To what extent can this be considered true?

It's utter bullfrogs.

[zompist]

...the universal constraint that plurals may not show a greater degree of morphemic contrast than the corresponding singulars... (p105, #3)

Is this actually a thing, or does it not mean what I think it means, that morphology is always more complex in singulars than plurals? It's not hard to find counterexamples

I'm struggling to understand what he means. I assume, that case and gender distinctions will be fewer in plural declensions? I think that's broadly true of Indo-European. Well, unless you look at verbs— e.g. (spoken) French verbs distinguish person in the plural, generally don't do so in the singular.

[Darren]

Penny's referring to the loss of the nominative plural in feminine nouns in proto-Romance. Latin -A -AM -AE -ĀS should give proto-Romance *-a -a -e -as but instead all descendants, including those that retained case, reflect *-a -a -as -as. His theory is that this happened because having a case distinction in the plural but not in the singular is unstable, which might be at least a universal tendency. It's hard to see why else this would have been lost, unless perhaps rural dialects resisted the analogical introduction of -AE into the paradigm in place of proto-Italic *-ās.

[hwhatting]

...the universal constraint that plurals may not show a greater degree of morphemic contrast than the corresponding singulars... (p105, #3)

It's not a constraint, it is a tendency - normally expressed as something like "marked categories usually show less additional category distinctions than unmarked categories". But it's a tendency, to which individual counterexample can be easily found. Adding to what zompist said, a counterexample for nominal morphology is some noun declension classes in German that show no case distinction in the singular but have a separate case form for the dative plural.

And, a reputable source which I annoyingly can't remember says that Latin got articles from Greek, which got them from Egyptian. To what extent can this be considered true?

I assume they mean Vulgar Latin / Proto -Romance? I wouldn't exclude that Greek played a role, as much of Southern Italy was Greek-speaking and there must have been sizeable Greek minorities all over the empire. OTOH, Greek had definite articles long before Greeks and Egyptians came into close contacts, so I think that part is bullshit.

[Travis B.]

...the universal constraint that plurals may not show a greater degree of morphemic contrast than the corresponding singulars... (p105, #3)

It's not a constraint, it is a tendency - normally expressed as something like "marked categories usually show less additional category distinctions than unmarked categories". But it's a tendency, to which individual counterexample can be easily found. Adding to what zompist said, a counterexample for nominal morphology is some noun declension classes in German that show no case distinction in the singular but have a separate case form for the dative plural.

Your German example is quite the counterexample to this tendency, since only weak masculine nouns (for those who don't know German well, nouns such as Herr, which is Herrn in the dative singular) and some rather dated/set expression noun forms (such as Hause, found today primarily in the set expressions zu Hause and nach Hause) distinguish the dative singular from the nominative singular at all on the noun itself, yet it is extremely common for nouns to distinguish the dative plural from the nominative plural on the noun itself in StG.

And, a reputable source which I annoyingly can't remember says that Latin got articles from Greek, which got them from Egyptian. To what extent can this be considered true?

I assume they mean Vulgar Latin / Proto -Romance? I wouldn't exclude that Greek played a role, as much of Southern Italy was Greek-speaking and there must have been sizeable Greek minorities all over the empire. OTOH, Greek had definite articles long before Greeks and Egyptians came into close contacts, so I think that part is bullshit.

People forget about Magna Graeca and the extent of Greek colonization long before the Hellenistic era (which occurred at a rather late date by Greek standards, and which was the era in which the Greeks came into significant contact with the Egyptians). Remember, the Romans got the alphabet from the Etruscans, who got it from Magna Graeca.