Useful things to know and think about for Language Policy

Schiffman's 'Laws' of Language Policy


  1. There is no such thing as 'no language policy' . If there doesn't seem to be an explicit language policy, the policy is implicit, covert, de facto, unwritten, customary, laissez-faire.

  2. Language policies, however explicit, are typically underspecified. That is, no matter how specific they are, they are never explicit enough to cover all contingencies. (See 'unintended consequences', below).

  3. Implementation. When language polices fail (or get into trouble), it is typically when it comes to implementation of the policy. Language policy planners typically fail to anticipate all the ramifications of implementation (costs, time, follow-up, cultural factors, demographic changes) and often act as if vaguely-worded policies, guided by a lot of wishful thinking, will somehow take care of themselves. They thus fail to deal with unintended consequences or unanticipated developments, or factors beyond their control. Language policies are (as someone put it in reference to other kinds of policies) not self-implementing. Another way of putting it is (as someone said) "Hope is not a policy." (More on the problems of implementation here.)

  4. Costs: Language policies have a cost, whether this be the financial costs of implementation, training of teachers, publication of teaching materials, verification and enforcement of the policy, testing, etc., or the typically unreckoned human costs (confusion, wasted human resources, inconvenience, suffering, alienation...). Polities often fail to balance costs with benefits, or they ignore certain costs or certain benefits. And they also often fail to follow through the entire calendar of implementation, budgeting for costs 5 or 10 years in the future. Or, regimes change, priorities change, and the commitment to paying for the costs fizzles.

  5. Self-implementing and Self-Perpetuating Policies. Policy planners tend to think a policy can be developed, set in stone, enshrined in law, and that the issue is then solved once and for all. (I.e., they act as if they are self-implementing, and in perpetuity.) Retrospectively, people act as if policies developed years or centuries ago will remain valid, and must be treated as if still valid (even if conditions have changed.) They fail to see the evolutionary aspects of policy, and that policies (like regimes, like populations, like attitudes), typically evolve and change (or that if they fail to evolve, that they will fail). We might refer to this (borrowing a term from discussions of our Supreme Court) as "strict constructionism." (By this way of thinking, the abolition of slavery was not part of the original intent of those who wrote the US Constitution, so we should still be maintaining it, despite amendments to the constitution since then.)


Harold F. Schiffman, haroldfs@ccat.sas.upenn.edu, last modified Jan. 12, 2006