This question and responses illustrate how easy it is to get sidetracked from the abstract moral question by the empirical difficulties involved in getting information in very complicated social situations.
It is extremely easy to answer the abstract moral question.
Can cruelty and suffering be justified if it might lead to social transformation, or to possibly establish an ideal state?
With a simple example. Suppose it were the case that simply pinching somewhat and causing them a very brief sensation of pain would result in freeing 10 million people from a situation of horrible slavery. We also suppose that the person to be pinched is completely innocent of any wrongdoing.
Would such a pinch be justified? It is cruel and causes some suffering.
Resounding YES! The answer to the question - YES ! - is beyond question. Any other answer is patently absurd.
So what's going on with all the discussion? If you look over the answers there are a great many points brought up. Points that are related as to whether in various particular circumstances the amount of suffering would justify the intended result and whether or not allowing the suffering would even have the intended result.
These are empirical matters which can be impossible to determine beforehand. What suffering is allowable will depend strongly on one's belief about what the outcome is likely to be.
Other points raised are whether or not the desired changes are actually desirable. Of course if two people disagree on whether or not the intended changes are desirable they won't agree on causing someone pain and suffering to achieve those ends. The question is quite ambiguous about this, it is asked as if there is a general answer
Can cruelty and suffering be justified if it might lead to social transformation, or to possibly establish an ideal state?
As such it is asking
Can cruelty and suffering ever be justified if it might lead to social transformation, or to possibly establish an ideal state?
At the other extreme we could have:
Is cruelty and suffering always justified if it might lead to social transformation, or to possibly establish an ideal state?
Other clarifications come to mind:
Can cruelty and suffering ever be justified if it almost certainly will lead to social transformation and establish an ideal state?
Is cruelty and suffering always justified if it almost certainly will lead to social transformation and establish an ideal state?
All these alternatives show that the question can only be meaningfully answered if the actual circumstances are known. What are the probable effects of allowing a specific amount and type of suffering. These empirical relations are of course extremely difficult to come by so oblique arguments are made instead.
It's really a version of asking "Do the ends justify the means." without ever saying what ends? what means? The question has become stripped of the context absolutely necessary for it's answer.