When flight instruments, especially the vacuum-dependent gyros (AH and DG) were introduced c. 1930 ±, the simplest, lightest-weight (including connections) source for motive power for the gyros was a venturi tube outside the airframe a few inches from the instrument panel, providing suction. "So simple even a Piper Cub could do it."
https://upload.wikimedia.org/wikiped..._venturi_1.JPG
Much lighter than a pump casing and pipes, that had to hold pressure, and reach from the engine through the firewall. And, of course, aircraft engineers
understood suction and vacuum (Bernoulli, Venturi, etc.).
Once vacuum drive became the standard - and worked - why fix something that ain't broke?
Remind your friend that in
aircraft engineering, the first question in planning a new system is not "What's the
best way we can do it?", but always "What's the
lightest, smallest way we can do it?"
Eventually, the trade-off of high-speed drag vs. weight moved the suction source to the manifold from the external venturis. But the manifold suction was already there - free, gratis and fer nuttin' - compared to adding a dedicated, relatively heavy accesory pump. Plus gauges were already designed around "suck" and not "blow," and why redesign them?