Ah, less vs. fewer. Another arrow in the prescriptivist’s quiver of pointless pedantry.
There's even a Wikipedia article about the dispute. There is also a Language Log entry about the matter too.
According to Merriam-Webster's Dictionary of English Usage, a usage guide that looks carefully at the history of usage advice, the rule creating a clear separation for less and fewer was invented in 1770 by Rober Baker in his book Reflections on the English Language, where he wrote in a comment on less:
The Word is most commonly used in speaking of a Number; where I should think Fewer would do better. No Fewer than a Hundred appears to me not only more elegant than No less than a Hundred, but more strictly proper.
The Merriam-Webster Dictionary of English Usage authors then comment:
Baker’s remarks about fewer express clearly and modestly—“I should think,”, “appears to me”—his own taste and preference. It is instructive to compare Baker with one of the most recent college handbooks in our collection:
Fewer refers to quantities that can be counted individually.… Less is used for collective quantities that are not counted individually… and for abstract characteristics. —Trimmer & McCrimmon 1988
Notice how Baker’s preference has here been generalized and elevated to an absolute status, and his notice of contrary usage has been omitted. This approach is quite common in handbooks and schoolbooks; many pedagogues seem reluctant to share the often complicated facts about English with their students.
How Baker’s opinion came to be an inviolable rule, we do not know. But we do know that many people believe it is such.
They then give many examples of usage of less for countable quantities, and add finally:
The examples above show native speakers and writers of English using less of count nouns in various constructions. Fewer could have been used in many of them—at times it might have been more elegant, as Robert Baker thought—but in others no native speaker would use anything but less.
With regards to the example in the original question, either fewer or less would be perfectly grammatical, but so many people are under the spell of the rule that less must never be used with countable nouns that anyone who doesn’t follow the rule may be subject to criticism.
Edit 2010-09-28:
The Cambridge Grammar of the English Language also weighs in on less vs. fewer:
The relation between less and fewer is fairly complex. In non-count singulars only less is possible: Kim has less/*fewer money than Pat. In plural NPs we have:
[17]
i. She left less than ten minutes ago.
ii. Less/Fewer than thirty of the students had voted.
iii. He made no less/fewer than fifteen mistakes.
iv. You pass if you make ten mistakes or less/?fewer.
v. He took less/*fewer pains to convince us than I’d expected.
vi. He made fewer/less mistakes than the others.
Both [i] and [ii] have than + numeral. In [i] ten minutes expresses an amount of time rather than a number of individuated units, and in such cases fewer is virtually impossible—just as few would be in a comparison of equality: She left as little/*few as ten minutes ago. Similarly with We paid less than thirty dollars for it; She’s less than forty years old; We were going at less than ten miles an hour. In [ii] we are concerned with countable individuals and little cannot be used in a comparison of equality (*as little as thirty of the students); nevertheless, for inequality less is more common than fewer in this construction. The same applies with percentages: Less/Fewer than 30% of the students had voted. Construction [iii] has the comparative form following no: though the interpretation is count plural, less is here again more common than fewer. Construction [iv] has or after a numeral: less is the usual form here, with fewer quite marginal; this construction is widely seen in supermarkets, with the fast checkout labelled eight items or less, or the like. In [v] pains is plural but non-count rather than count (we can’t ask how many pains he took), and here only less is possible. Finally in [vi] (as also in [15ii]) the comparative occurs directly with a count plural noun: both forms are found, but less is subject to quite strong prescriptavist disapproval, so that fewer is widely preferred in formal style, and many speakers in informal style too.
[Usage manuals are divided on the issue of less vs. fewer. Some uncompromisingly brand such forms as less mistakes as incorrect, while others note that though commonly condemned they are often used by speakers of Standard English. Before the Early Modern English period (beginning around 1500) more was restricted to non-count NPs with moe used as the comparative of many. At that time less was used along with fewer for count NPs, but came to be stigmatised and quite rare in this use: it is only within the last generation or so that it has become frequent. The current revival seems inexorable, given the strong pressure of analogy with more.]
Best Answer
I think that how you choose to frame the comparison depends to a large extent on how granularly you define you define your categories.
Suppose that Nation A permits unrestricted movement on the part of its citizens by public transport (bus, train, or airplane) but bans private ownership and use of cars, motorcycles, and bicycles. And suppose that Nation B permits unrestricted movement on the part of its citizens by public and private modes of transport.
At a generalized level, both nations give their citizens a "freedom to travel," so it is incorrect (at that level of generality) to say that Nation A has fewer freedoms than Nation B: Both nations can check off the box for "freedom to travel," and both are tied at one freedom each.
But if you itemize the particular ways of traveling, and treat each one as a potential freedom—"freedom to travel by bus," "freedom to travel by train," "freedom to travel by airplane," "freedom to travel by private car," "freedom to travel by private motorcycle," and "freedom to travel by private bicycle"—Nation A suddenly loses the freedoms competition, six to three. Nation A has fewer freedoms (in the obscure corner of societal freedoms that we're talking about here) than Nation B does.
You might also conclude, at this level of analysis, that Nation A has less freedom with regard to travel than Nation B does, because Nation A has prohibited an area of action by citizens that is (formally, at least) unrestricted in Nation B.
Of course, many complications may arise. For example, suppose that in Nation A there is no fee for taking public transportation within the country's borders, and that for many citizens in Nation B public transportation fees are discouragingly high and private ownership of cars, motorcycles, and bicycles is prohibitively expensive. Then we might say that although Nation B offers its citizens more freedoms than Nation A does, a poor person in Nation B actually has less practical freedom to travel than does a person of any economic class in Nation A.
Part of the trouble here has to do with addressing the issue of freedom and freedoms as though these words describe something that a nation possesses, rather than as something that its citizens possess. And another part of the trouble is that a chasm frequently exists between a person's abstract right to something and a person's practical possession of it.
To circle back to your original question, it seems to me that both the notion of "more and fewer freedoms" and the notion of "more and less freedom" are quite coherent and distinct ideas that may be discussed meaningfully under suitable conditions. But they are also extremely slippery ideas because they can be appraised very differently at different levels of generality or granularity, at different gradations of abstractness and concreteness, and with regard to different populations within the state or nation.