15.  Having devoted many years to studies of this nature, and being conversant with most of the grammatical treatises already published, the author conceived that the objects above referred to, might be better effected than they had been in any work within his knowledge.  And he persuades himself, that, however this work may yet fall short of possible completeness, the improvements here offered are neither few nor inconsiderable.  He does not mean to conceal in any degree his obligations to others, or to indulge in censure without discrimination.  He has no disposition to depreciate the labours, or to detract from the merits, of those who have written ably upon this topic.  He has studiously endeavoured to avail himself of all the light they have thrown upon the subject.  With a view to further improvements in the science, he has also resorted to the original sources of grammatical knowledge, and has not only critically considered what he has seen or heard of our vernacular tongue, but has sought with some diligence the analogies of speech in the structure of several other languages.  If, therefore, the work now furnished be thought worthy of preference, as exhibiting the best method of teaching grammar; he trusts it will be because it deviates least from sound doctrine, while, by fair criticism upon others, it best supplies the means of choosing judiciously.

16.  Of all methods of teaching grammar, that which has come nearest to what is recommended above, has doubtless been the most successful; and whatever objections may have been raised against it, it will probably be found on examination to be the most analogous to nature.  It is analytic in respect to the doctrines of grammar, synthetic in respect to the practice, and logical in respect to both.  It assumes the language as an object which the learner is capable of conceiving to be one whole; begins with the classification of all its words, according to certain grand differences which make the several parts of speech; then proceeds to divide further, according to specific differences and qualities, till all the classes, properties, and relations, of the words in any intelligible sentence, become obvious and determinate:  and he to whom these things are known, so that he can see at a glance what is the construction of each word, and whether it is right or not, is a good grammarian.  The disposition of the human mind to generalize the objects of thought, and to follow broad analogies in the use of words, discovers itself early, and seems to be an inherent principle of our nature.  Hence, in the language of children and illiterate people, many words are regularly inflected even in opposition to the most common usage.