minister both can be and ought to be responsible, and the sphere of
ordinary administrative work for which the minister cannot be genuinely
responsible? If that distinction is accepted, it ought not to be
impossible for Parliament without undermining ministerial or cabinet
responsibility, to devise a means of making its control over the
ordinary working of the departments effective, through a system of
committees or in other ways.

The current complaints of bureaucracy, however, are not directed mainly
against the ineffectiveness of the machinery of control, but against the
way in which public work is conducted by government officials--the
formalism and red-tape by which it is hampered, the absence of
elasticity and enterprise; and the methods of government departments are
often compared, to their disadvantage, with those of business firms. But
the comparison disregards a vital fact. The primary function of a
government department is not creative or productive, but regulative. It
has to see that laws are exactly carried out, and that public funds are
used for the precise purposes for which they were voted; and for this
kind of work a good deal of red-tape is necessary. Moreover, it is
essential that those who are charged with such functions should be above
all suspicion of being influenced by fear or favour or the desire to
make profit; and for this purpose fixed salaries and security of tenure
are essential.

In short, the fundamental principles upon which government departments
are organised are right for the regulative functions which they
primarily exist to perform. But they are altogether wrong for creative
and productive work, which demands the utmost elasticity, adaptability,
and freedom for experiment. And it is just because the ordinary
machinery of government has been used on a large scale for this kind of