I think the argument is probably:
0.) some people have been historically advantaged over each other for non-meritocratic reasons.
1.) one of the ways people have been advantaged is itself in the notability of their actions.
[0]: http://listverse.com/2013/10/14/10-groundbreaking-women-scie...
[1]: http://io9.gizmodo.com/5077952/women-who-pretended-to-be-men...
[2]: https://www.wsj.com/articles/SB10001424127887324355904578159...
[3]: https://fivethirtyeight.com/datalab/are-female-scientists-hi...
2.) Therefore, using notability as a prerequisite for inclusion perpetuates a non-meritocratic historical bias.
To your question: "Is there any guarantee that getting rid of the notability criteria is going to result in a more meritocratic system?" I think that there are no guarantees in cultural changes, and the best way to be wrong about the outcomes of cultural change is to predict any outcome with certainty. That said, if uncertainty was a sufficient argument against cultural change, then we'd all be hunter/gatherers.