According to the _Theory of Utility_, as warmly supported by Mr. Hume, we estimate the virtue of an action and an agent entirely by their usefulness. He seems to refer all our mental impressions to two principles, reason and taste. Reason gives us simply the knowledge of truth or falsehood, and is no motive of action. Taste gives an impression of pleasure or pain,--so constitutes happiness or misery, and becomes a motive of action. To this he refers our impressions of beauty and deformity, vice and virtue. He has, accordingly, distinctly asserted that the words right and wrong signify nothing more than sweet or sour, pleasant or painful, being only effects upon the mind of the spectator produced by the contemplation of certain conduct,--and this, as we have already seen, resolves itself into the impression of its usefulness. An obvious objection to the system of utility was, that it might be applied to the effects of inanimate matter as correctly as to the deeds of a voluntary agent. A printing-press or a steam-engine might be as meritorious as a man of extensive virtue. To obviate this, Mr. Hume was driven to a distinction, which in fact amounted to giving up the doctrine, namely, that the sense of utility must be combined with a feeling of approbation. This leads us back to the previous question, on what this feeling of approbation is founded, and at once recognises a principle, distinct from the mere perception of utility. Virtuous conduct may indeed always contribute to general utility, or general happiness, but this is an effect only, not the cause or the principle which constitutes it virtuous. This important distinction has been well stated by Professor Mills of Oxford. He defines morality to be,--"an obedience to the law and constitution of man's nature, assigned him by the Deity in conformity to his own essential and unchangeable attributes, the effect of which is the general happiness of his creatures."[2]--We may safely assert, that whatever is right is also expedient for man; but the converse by no means follows,--that what is expedient, that is what mankind think would be expedient, comes to be right. [Footnote 2: Lecture on the Theory of Moral Obligation. Oxford, 1830.]