For the quick reader, find a summary of my argument at the bottom of this post.
There are basically two (classically liberal) approaches to liberty, one based on the harm principle, while the other approach admits to its premises the benefit principle in addition to the harm principle.
Liberty and the harm principle
Richard Epstein characterises the harm principle thus:
In setting out individual rights and duties, we must embrace principles of individual autonomy, private property, and voluntary exchanges in order to insulate these productive human activities from the ravages of force and fraud. (Principles for a Free Society, p. 320)
The locus classicus is found in Mill:
That principle is, that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. That the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right. These are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him, or visiting him with any evil in case he do otherwise. To justify that, the conduct from which it is desired to deter him must be calculated to produce evil to some one else. The only part of the conduct of any one, for which he is amenable to society, is that which concerns others. In the part which merely concerns himself, his independence is, of right, absolute. Over himself, over is own body and mind, the individual is sovereign.
John Stuart Mill (1859): On Liberty, chapter 1, paragraph 9.
As Richard Epstein notes, the conception of freedom based on the harm principle alone
has often been seen as treating force and fraud as the only problems worthy of a collective legal response, and thus has been frequently attacked as ignoring the wide range of holdout, coordination, and networking problems that arise in any complex society.
Incidentally, I do not think that a strict application of the harm principle is feasible, nor would its best approximation really achieve the aims of liberty, for it would create a situation of anarchy, where everyone has to fend for himself, which corresponds to a very low level of human development with hardly any division of labour. There will be a lot of anthropocentric freedom favouring the roaming or the stationary bandits (of Mancur Olson's primordial state) that soon tend to dominate such a primitive anarchist environment, but no/little sociogenic freedom, which latter we, classical liberals, have in mind, when we speak of liberty.
Liberty and the benefit principle
At any rate, to improve on the deficits of a concept of liberty solely based on the harm principle, classical liberals have come up with what Epstein describes as
a more complete theory of laissez-faire which acknowledges the need for legal rules that forthrightly govern both common property and forced exchanges. In those cases where voluntary exchanges cannot achieve potential widespread gains, public force may take up the slack to achieve the desired social outcome - the win/win situations not obtainable by private agreement.
As the benefit principle involves coercion in that it will be applied in the face of non-concurrence, it may be more precisely defined as the principle whereby - under certain conditions - one should receive compensation for benefits conferred on others without their consent.
Liberty and the harm principle plus the benefit principle
Now, Epstein offers a crucial observation, which to my mind, indicates both the important task of "freedom as method", as well as the fundamental incompleteness of freedom, which forces us to figure out creatively and competitively solutions that liberty does not carry implied within her premises and principles.
Accepting that [benefit-]principle does not clear the path for the promiscuous use of state power. Rather it requires some clear showing that the individuals subjected to state power all benefit on net from the program that has taken or regulated their property.
The benefit principle as such does neither necessarily lead to egregious miscarriage of state power nor does it of itself protect us against such abuse. The reason why is that
some clear showing that the individuals subjected to state power all benefit on net from the program that has taken or regulated their property
is mostly, and in many crucial cases, not possible. The competitive efforts at a "clear showing" are manifold, amounting to a dramatic show of multitudinous dissent and Babylonian confusion in their own right. There is no single argument, theory, or philosophy that can tell you what the common weal is: "the benefit on net." We must conspire to huddle more or less comfortably under the umbrella of a rickety pretense of knowing what that common good might be that persuades us not to cut each others throats. That is the very best we can do.
This marks the most momentous disagreement that I have with Richard Epstein, whom I admire greatly for his classical liberal reconstruction of law and the Constitution. As liberals, I am suggesting, we can make plausible proposals as to what constitutes the common weal ("freedom as method"), but we are no more capable of proving our point of view as objectively valid than does any dissenting party.
We, the citizens of a country, may end up very broadly in agreement in that we accept by and large actions carried out under the benefit principle, but this can not be based on an accurate reckoning concerning each issue or even the overall proportionality of state coercion/taking and benefit by the people.
Politics - the fuzzy logic of trust
Of course, where possible at all, we should strive toward such accurate reckoning (which striving is an important function of what I call "freedom as method"), but I dare say, in the final analysis
what makes for reasonable social cohesion in a free society with free access to the processes of politics for all (who care to take the trouble of getting involved), is the degree of trust that the opposing partisans ultimately place in one another, strong conflict of opinion notwithstanding.
A trust that is rarely expressed explicitly, a trust that the trusting are often not even aware of, a trust that happens even when partisans think nothing good of one anther. A trust that shows in the fact that people do participate (at great cost and with no payback to them) in the political game of a country.
And that trust is not anything of mathematical or logical precision (that the pretension of an axiomatic, all-knowing theory of liberty aspires to); it is a cultural event with thousands of facets to it, one of the most important of which is political participation, in which we get to practice tolerance and respect for the opponent, a political culture of compromise, as well as select and sagaciously chosen bipartisanship.
Summary: Political participation of those conscious of the need and conditions of freedom is indispensable. Freedom is based on the adherence of certain principles; however, from the principles of freedom - notably the harm principle and the benefit principle - we cannot derive unequivocal and generally accepted answers to all the challenges that a human community is confronted with. At the end of the day, liberty is an open-ended system that still leaves us with the need to compete, negotiate, and compromise politically. There must be space for those whose views deviate from ours. Good politics is about such accommodation. For it is impossible to build a world that is 100% in accordance with principles of freedom, simply because it is impossible to arrive at a uniform, uncontested, unanimously supported concept of freedom. In the face, of a plurality of views, and with people having the right to disagree with one another - one of the most basic principles of freedom -, liberty cannot be an arsenal of forgone conclusions, but becomes a method and attitude to be applied to the open issues that we are facing in the contemporary political debate. The skillful politician asks the right critical questions that introduce our concern for liberty into the debates and decisions of politics, and she is at the same time capable of accommodating views that arise in areas where there is no unique liberty-minded answer available. And she must, at times, accommodate even decisions that she is not happy about at all from a principled and truthful point of view. After all, politics is also about creating viable trust in a community, and to that purpose it is necessary to let the other side prevail, so that my side can prevail, so that the other side can prevail, so that my side can prevail ... This difficult, yet absolutely necessary task is the job of the politician, for which she will be always scolded by some, even if she handles her calling with the utmost responsibility. Good politics is about doing and achieving the imperfect so that the community can thrive.