Frank Jackson's 'Mary the Color Scientist' is criticized because its central premise—that someone could know 'all the physical facts' about color—is fundamentally inconceivable. This makes it difficult to draw any reliable intuitive conclusions from the scenario.
The classic thought experiment is deemed ineffective because it doesn't genuinely probe a moral intuition. Instead, it merely restates the argument's premise—that people are only moral for instrumental reasons—packaging a conclusion as a speculative scenario.
The famous thought experiment, while popular, proved detrimental to moral psychology and ethics. It narrowed inquiry and promoted a forced, simplistic choice between consequentialist and deontological reasoning, stifling more nuanced understanding of the human mind.
The initial power of a thought experiment lies in its ability to provoke a novel intuition. However, this value is often quickly drained by subsequent academic debates that over-analyze and dissect it, stripping it of its original interest.
J.J. Thompson's 'famous violinist' example is judged effective not as a broad defense of abortion, but for its success in making a very narrow point: that the right to life is not absolute, even if one grants personhood.
Nozick's thought experiment is ineffective because it doesn't engage with common-sense moral intuitions. It's a problem that only arises if one is already deeply committed to a specific, extreme utilitarian framework, making it feel absurd and irrelevant to most people.
While the thought experiment made academic political philosophy overly abstract and detached from reality ('ideal theory'), it remains a valuable pedagogical tool for compelling students to consider societal justice from the perspective of the least advantaged.
Elite thought experiments like Singer's Drowning Child are powerful because their scenarios are relatable and don't require suspending disbelief about complex conditions, unlike variations of the Trolley Problem that rely on a 'fat man' perfectly stopping a train.