were subject to the will, is it not natural to conclude, that in the one case, as well as in the other, we would endeavour to banish, as much as we could, every idea which had a tendency to disturb us, and detain those only which we found to be agreeable? So far, however, is this power over our thoughts from being exercised, that we are frequently oppressed, in spite of all our efforts to the contrary, with dreams which affect us with the most painful emotions. And, indeed, it is matter of vulgar remark, that our dreams are, in every case, involuntary on our part, and that they appear to be obtruded on us by some external cause. This fact appeared so unaccountable to the late Mr. Baxter, that it gave rise to his very whimsical theory, in which he ascribes dreams to the immediate influence of separate spirits on the mind.

3. If the influence of the will be suspended during sleep, the conceptions which we then form of sensible objects, will be attended with a belief of their real existence, as much as the perception of the same objects is while we are awake.

In treating of the power of Conception, I formerly observed, that our belief of the separate and independent existence of the objects of our perceptions is the result of experience, which teaches us that these perceptions do not depend on our will. If I open my eyes, I cannot prevent myself from seeing the prospect before me. The case is different with respect to our conceptions. While they occupy the mind, to the exclusion of everything else, I endeavoured to show that they are always accompanied with belief; but as we can banish them from the mind, during our waking hours, at pleasure, and as the momentary belief which they produce is continually checked. by the surrounding objects of our perceptions, we learn to consider them as fictions of our own creation, and, excepting in some accidental cases, pay no regard to them in the conduct of life. If the doctrine, however, formerly stated with respect to conception be just, and if, at the same time, it be allowed that sleep suspends the influence of the will over the train of our thoughts, we should naturally be led to expect, that the same belief which accompanies perception while we are awake, should

accompany the conceptions which occur to us in our dreams. It is scarcely necessary for me to remark how strikingly this conclusion coincides with acknowledged facts.

May it not be considered as some confirmation of the foregoing doctrine, that when opium fails in producing complete sleep, it commonly produces one of the effects of sleep, by suspending the activity of the mind, and throwing it into a reverie; and that while we are in this state, our conceptions frequently affect us nearly in the same manner as if the objects conceived were present to our senses ?1

Another circumstance with respect to our conceptions during sleep, deserves our notice. As the subjects which we then think upon occupy the mind exclusively, and as the attention. is not diverted by the objects of our external senses, our conceptions must be proportionably lively and steady. Every person knows how faint the conception is which we form of any thing with our eyes open, in comparison of what we can form with our eyes shut; and that in proportion as we can suspend the exercise of all our other senses, the liveliness of our conception increases. To this cause is to be ascribed, in part, the effect which the dread of spirits in the dark has on some persons, who are fully convinced in speculation that their apprehensions are groundless; and to this also is owing the effect of any accidental perception in giving them a momentary relief from their terrors. Hence the remedy which nature points out to us, when we find ourselves overpowered by imagination. If every thing around us be silent, we endeavour to create a noise, by speaking aloud, or beating with our feet; that is, we strive to divert the attention from the subjects of our imagination, by presenting an object to our powers of perception. The conclusion which I draw from these observations is, that as there is no state of the body in which our perceptive powers are so totally unemployed as in sleep, it is natural to think that the objects which we conceive or imagine must then make an impression on the mind, beyond comparison greater than any thing of which we can have experience while awake.

See the Baron de Tott's Account of the Opium-takes at Constantinople.

From these principles may be derived a simple, and I think a satisfactory explanation of what some writers have represented as the most mysterious of all the circumstances connected with dreaming; the inaccurate estimates we are apt to form of Time, while we are thus employed;-an inaccuracy which sometimes extends so far, as to give to a single instant the appearance of hours, or perhaps of days. A sudden noise, for example, suggests a dream connected with that perception, and the moment afterwards this noise has the effect of awaking us; and yet during that momentary interval a long series of circumstances has passed before the imagination. The story quoted by Mr. Addison1 from the Turkish tales, of the miracle wrought by a Mahometan doctor to convince an infidel sultan, is in such cases nearly verified.

The facts I allude to at present are generally explained by supposing, that in our dreams the rapidity of thought is greater than while we are awake; but there is no necessity for having recourse to such a supposition. The rapidity of thought is at all times such, that in the twinkling of an eye a crowd of ideas may pass before us, to which it would require a long discourse to give utterance; and transactions may be conceived which it would require days to realize. But in sleep the conceptions of the mind are mistaken for realities; and therefore our estimates of time will be formed, not according to our experience of the rapidity of thought, but according to our experience of the time requisite for realizing what we conceive. Something perfectly analogous to this may be remarked in the perceptions we obtain by the sense of sight. When I look into a show-box where the deception is imperfect, I see only a set of paltry daubings of a few inches' diameter; but if the representation be executed with so much skill as to convey to me the idea of a distant prospect, every object before me swells in its dimensions in proportion to the extent of space which I conceive it to occupy; and what seemed before to be shut up within the limits of a small wooden frame, is magnified, in my apprehension, to an immense landscape of woods, rivers, and mountains. 1 Spectator, No. 94.

The phenomena which we have hitherto explained, take place when sleep seems to be complete; that is, when the mind loses its influence over all those powers whose exercise depends on its will. There are, however, many cases in which sleep seems to be partial; that is, when the mind loses its influence over some powers, and retains it over others. In the case of the somnambuli, it retains its power over the limbs, but it possesses no influence over its own thoughts, and scarcely any over the body, excepting those particular members of it which are employed in walking. In madness, the power of the will over the body remains undiminished, while its influence in regulating the train of thought is in a great measure suspended, either in consequence of a particular idea which engrosses the attention to the exclusion of every thing else, and which we find it impossible to banish by our efforts, or in consequence of our thoughts succeeding each other with such rapidity, that we are unable to stop the train. In both of these kinds of madness it is worthy of remark, that the conceptions or imaginations of the mind becoming independent of our will, they are apt to be mistaken for actual perceptions, and to affect us in the same

manner.

By means of this supposition of a partial sleep, any apparent exceptions which the history of dreams may afford to the general principles already stated, admit of an easy explanation.

Upon reviewing the foregoing observations, it does not occur to me that I have in any instance transgressed those rules of philosophizing, which, since the time of Newton, are commonly appealed to as the tests of sound investigation. For, in the first place, I have not supposed any causes which are not known to exist; and, secondly, I have shewn that the phenomena under our consideration are necessary consequences of the causes to which I have referred them. I have not supposed that the mind acquires in sleep any new faculty of which we are not conscious while awake, but only (what we know to be a fact) that it retains some of its powers, while the exercise of others is suspended; and I have deduced synthetically the known phenomena of dreaming, from the operation of a par

ticular class of our faculties, uncorrected by the operation of another. I flatter myself, therefore, that this inquiry will not only throw some light on the state of the mind in sleep, but that it will have a tendency to illustrate the mutual adaptation and subserviency which exist among the different parts of our constitution, when we are in complete possession of all the faculties and principles which belong to our nature.1

PART SECOND.

OF THE INFLUENCE OF ASSOCIATION ON THE INTELLECTUAL AND ON THE ACTIVE POWERS.

SECT. I.-OF THE INFLUENCE OF CASUAL ASSOCIATIONS ON OUR
SPECULATIVE CONCLUSIONS.

The association of ideas has a tendency to warp our speculative opinions chiefly in the three following ways:—

First, by blending together in our apprehensions, things which are really distinct in their nature, so as to introduce. perplexity and error into every process of reasoning in which they are involved.

Secondly, by misleading us in those anticipations of the future from the past, which our constitution disposes us to form, and which are the great foundation of our conduct in life.

Thirdly, by connecting in the mind erroneous opinions, with truths which irresistibly command our assent, and which we feel to be of importance to human happiness.

A short illustration of these remarks will throw light on the origin of various prejudices, and may perhaps suggest some practical hints with respect to the conduct of the understanding.

I. I formerly had occasion to mention several instances of

VOL. II.

1 See Note O.

U