Just my thinking, anytime astronomy invokes multiple generations of gas clouds forming and reforming to explain stars, clusters, and globular clusters, falls outside the scientific method to verify as true. There are many reports on the how to explain the origin of globular clusters, GCs.
Example, "A new computer model explores star formation and uncovered that globular clusters initially
formed with many more massive stars than in clusters observed today. "ScienceDaily (May
11, 2012)", How Nature Shapes the Birth of Stars,
How nature shapes the birth of stars -- ScienceDaily
May-2021.
"A new computer model suggests globular clusters (GC) formed from galaxy collisions
including the nearly 200 in the Milky Way. This model shows many small clusters of stars
do not form but only the larger clusters remain after galaxy collisions which should take place
some 11-12 billion years ago in the big bang model.", Globular Clusters: Survivors of a 13-Billion-Year-Old Massacre,
Globular clusters: Survivors of a 13-billion-year-old massacre -- ScienceDaily , Feb-2012
The Formation of Globular Clusters as a Case of Overcooling,
https://ui.adsabs.harvard.edu/abs/2022arXiv220303002R/abstract, March-2022.
“3 WHERE DID GLOBULAR CLUSTERS FORM? The GC with the most complex display of multiple populations is undoubtedly ω Centauri, which hosts at least 15 distinct stellar populations (Bellini et al. 2017). Though at that time such complexity was not known yet, Bekki & Freeman (2003) proposed that ω Cen was the remnant nuclear cluster of a now dissolved dwarf galaxy. We now wish to endorse an Ansatz such that not just ω Cen, but virtually all GCs formed inside dwarf galaxies (as first proposed by Searle & Zinn 1978), perhaps with the exception of the more metal rich ones. Indeed, the association of GCs with dwarf galaxies (as in the case of M54 and Sagittarius) and with Gaia streams is now being widely pursued (e.g., Massari, Koppelman & Helmi 2019). On the other hand, even the most metal poor GCs had to form in an environment in which the metallicity had reached about 1/100 solar, as expected for dwarf galaxies at over 12.5 Gyr lookback time, or z>∼ 3. In this picture GCs did not hurry to separate themselves from the womb where they formed. This may have taken several Gyr to do so. While waiting for the Galaxy to grow, young GCs were still immersed in the dense gas of the host dwarf that worked as a tamper holding the stellar ejecta within the new born cluster and possibly feeding it with stellar ejecta from stars in the dwarf itself. In any event, GCs are not pristine, though at least the metal poor largely predate the Galaxy. Often GCs are thought as having been accreted by the Galaxy, as if the Galaxy was already in place and the GCs arrived later, whereas it may well be that the actual late-comer is the bulk of the Galaxy itself, that was slowly built up by inflowing gas streams as currently understood (Dekel et al. 2009). Thus, as the Galaxy grew up, most dwarfs, but not all, have been tidally destroyed, then releasing their GCs. GCs are extremely compact objects, with density of the order of ∼ 10^7 particles cm^−1 (now all in stars), i.e., many thousand times that of typical molecular clouds in today’s Milky Way. Runaway gas cooling and gravitational collapse must have occurred to form them, but it is less clear how gas was squeezed to such high densities before forming stars. On the other hand, besides being very dense, GCs are also extremely slow rotators (Sollima, Baumgardt & Hilker 2019). It is possible that GCs can only form in deep local minima of the potential well created by the collapse of giant molecular clouds and in minima of the inter stellar medium (ISM) vorticity, which may explain why (thin) disk GCs do not exist. Otherwise, a mechanism should be invented as to remove angular momentum with extremely high efficiency from giant molecular clouds.”
It is clear from many papers published that showing how GCs formed is still very much, a work in progress
When I view M42 using my 10-inch telescope and M13, I ponder issues like this
So, these papers show GCs could form by galaxy collisions and mergers, very massive stars not seen today in the GCs, and giant gas clouds assumed to exist but still difficult to show how these form GCs because of density issues. All of these mechanisms cannot be seen when examining the 150 or more GCs in the MW today or other star clusters, e.g. M45 in Taurus.