O N M E A S U R E S O F V A L U E .
WE now come to the consideration of a subject which has made a conspicuous figure in the writings of political economists, and than which, none perhaps has been a greater source of error and confusion; I mean the measurement of value.
Our first inquiry must therefore be directed to the signification of the term. The analogies suggested by the word measure seem to have bewildered almost every author who has touched on the subject. It has been taken for granted that we measure value as we measure extension, or ascertain weight; and it has been consequently imagined, that to perform the opera-tion we must possess an object of invariable value.
Let us examine, therefore, how far measuring value and measuring space are similar operations. In every case of measuring we merely ascertain ratios the ratio which one thing bears to another. In measuring the length of an object we find what ratio it bears to the length of some other object, or in other words, how many times one is contained in the other. We measure the longitudinal extension of a piece of timber, for example, by a foot rule; that is, we find bow often the length of the latter is contained in the former, and this is effected by the actual application of the rule to the timber. It is a physical operation, by which we obtain the knowledge of a fact before unknown, the ratio of length subsisting between the object and the instrument we employ.
In measuring value, what resemblance to this operation can possibly be discovered? We may place two objects by the side of each other, or apply one to the other in any way we please,  but we shall never be able by such means to discover the relation of value existing between them. We shall never extort from them a single fact with which we were before unacquainted. What then is it possible to do in the way of measuring value? What kind of measurement is intended, when the term is so frequently employed? All that is practicable appears to be simply this : if I know the, value of A in relation to B, and the value of B in relation to C, I can tell the value of A and C in relation to each other, and consequently their comparative power in purchasing all other commodities. This is an operation obviously bearing no resemblance at all to the process of measuring length. There is no unknown fact discovered by a physical operation: it is in truth a calculation from certain data, a mere question in arithmetic. It is not, let it be observed, what on a first glance it may appear, like ascertaining the comparative length of two pieces of timber which cannot be brought into juxta-position, by means of a foot rule or other instrument which we apply  first to one and then to the other: it is far from being so much as this: it is merely like calculating the ratio of length between the two pieces of timber, after we are informed how many feet are contained in each. For of each commodity A and C the value in relation to B must be given, or, in other words, their value must be expressed in a common denomination, before their mutual relation can be ascertained ; just as in the case supposed the relation of each piece of timber to the foot rule must be given, before their relation to each other can be deduced. The actual application of the foot rule is that part of the process which is alone entitled to the appellation of measuring, the rest being mere calculation, but to this there is nothing at all analogous in any possible attempt to ascertain value. The way in which the commodity B would be used, in the above instance, is in truth as a medium of comparison, not a measure, yet it is the only process which bears any analogy to measurement.
It appears, therefore, that all we can under  stand by a measure of value, is some commodity which would serve as a medium to ascertain the relation subsisting between two other commodities, that we had no means of bringing into direct comparison. Thus, if I wished to know the relation in exchange between corn and cloth, and there happened to be no instance of direct barter of one of these commodities for the other, I could acquire the desired information only by ascertaining their relations to a third commodity. Supposing this commodity to be money, if a yard of cloth were worth 10s., and a bushel of corn 5s., I should learn immediately that a yard of cloth was worth two bushels of corn, and would have an equal power of commanding all other things in exchange, silver in this instance being the commodity employed as a measure. This kind of measure of value, which is merely a medium of comparison, and obviously quite dissimilar to a measure of length, is the only one which it is possible to have; and although money is the measure generally employed, and by far the  most convenient of all, yet any other commodity might answer the purpose.
Such a measure as this, however, has not contented political economists; it is only, they say, a measure of commodities at the same time: they have wished for something to measure the value of commodities at different periods.
Let us see what this amounts to: if it is wished to measure or compare the value of corn and cloth at one period with their value at another period, money will evidently answer the purpose. We have only to inquire the prices of corn and cloth at each period, and we shall then be able to ascertain how they have varied relatively to each other. If, in the year 1600, cloth was 20s. a yard, and corn 10s. a bushel, and in the year 1800, cloth was 10s. and corn 10s., then it would manifestly appear, that in 1600 a yard of cloth would command in exchange or be worth two bushels of corn, and in 1800 only one bushel. Thus by inquiring the prices of the commodities we should ascertain  their variations in value with regard to each other, and money would be the measure of value or medium of comparison which we employed. This is evidently using money for a measure of value in the same manner as in the first case, the only difference being that we apply it to two periods, and make a subsequent comparison of the results obtained from each.
We have therefore not yet arrived at the sense in which the term is employed by economists, who are desirous of measuring the value of commodities at different periods. They do not wish to compare the mutual value of two commodities, or the relation subsisting between two commodities at one period, with the relation subsisting between them at another period, for this would be effected by a simple reference to their prices. They state their object to be, to find some standard commodity by which they might measure the value of the same object A, at two or more different periods, or, in other words, its fluctuations in value.
 But in relation to what object is it wished to measure the value of A and its fluctuations ? We cannot speak of value, as I have before shown, without meaning value in something, and as only A and the standard commodity which may be called Bare here in question, the value of A must mean its value in B. It is wished therefore to measure the relation between A and B at two different periods by B, Which if it has any signification must imply, that it is wished to ascertain the value of A and B relatively to each other at two different periods. These are historical facts, and when we have learned them as we learn other facts, we shall certainly know the fluctuations which the relation between A and B has undergone ; but B is, in this procedure, by no means a measure of value, or a medium of comparison, any more than A. In a word, turn the matter as we please, we shall find that we can have no measure of value but in the sense already explained. From this examination it appears, that a  measure of value can mean nothing but a commodity employed as a medium of comparison, and that so far from its being impossible to have any thing perfectly capable of performing this function, we are in the daily use of one possessing all the perfection which it is possible to conceive. In regard to measuring or comparing value, there is no operation which can be intelligibly described or consistently imagined, but may be performed by the media of which we are in possession.
It is astonishing, indeed, to find how slight are the analogies with which economists have contented themselves on this subject, and which have served to preclude any close investigation of processes essentially different, although confounded under the same appellation. One of the most striking instances of this carelessness of examination is the notion of its being necessary, that a commodity should possess invariable value, in order to form a perfect measure of value; a notion which has passed unquestioned from one writer to another, and been  adopted without any suspicion of the false analogy and fundamental misconception on which it proceeds *. It is therefore essentially requisite, for a clear understanding of the present subject, to bring this opinion to the test of a close and minute examination.
The utter absurdity, however, of supposing, that a commodity to constitute a measure must be of invariable value, requires no deep research; it lies almost on the surface, and presents itself in numerous different aspects.
Invariable value must signify, as we have repeatedly shown, invariable in relation to some one or more commodities. Suppose A to be the commodity selected as a measure, and that it is invariable in value to B. I have here got an invariable value, but in what way am I to use it in regard to other things? When I have an invariable space, or an unvarying distance between two points, I can apply it mediately or immediately to all other spaces or distances within my reach, and ascertain their  respective ratios to it: but the invariable relation of value between A and B can tell me nothing of the mutual value of C and D; or, to vary the language, the power which A has to command B, can tell me nothing of that which C has to command D. I do not in any sense measure the relation of value between two commodities, by that existing between two other commodities. Invariable value, therefore, can be of no service. The only meaning to be attached to the phrase measuring value, the only operation implied in it, is, as we have seen, that comparison of the values of two objects which we are enabled to make by their separate relations to a third, or, in other words, by having these values expressed in a common term or denomination. But the capability of expressing the values of commodities has nothing to do with the constancy of their values, either to each other or to the medium employed; neither has the capability of comparing these expressions of value any thing to do with it. Whether A is worth 4 B or 6 B, and whether C is worth 8 B or 12 B, are cir-cumstances which make no difference in the power of expressing the value of A and C in B, and certainly no difference in the power of comparing the value of A and C when expressed.
This supposition, of the necessity of invariable value in any commodity employed as a measure, proceeds, as I have already remarked, on a false analogy. Because a measure of space must be invariable in its length, a measure of value, it has been argued, must be invariable in its value *. To expose the fallacy of this inference, let us examine in the first place, what are the character and circumstances of that invariableness which is requisite for a measure of length. All that is required ap-pears to be this, that when we measure the length of two objects by a third object, the length of the, latter, or the instrument employed, must remain the same until it have been applied to both the objects which are to be measured; or if it vary, it must vary in a known degree. Suppose it is wished to ascertain the relation of length between two trees lying apart from each other on the ground. The only requisite for doing this is a staff, or rod, or any other instrument which shall continue of the same length during the process of measuring. The process over, although the rod might be instantly altered in length, it would be as good a measure as before of the length of these or any other objects: for suppose the measurement to be repeated after this alteration in the instrument, the same relation of length between the two trees would be obtained. But if the rod varied in its length in an unknown degree, between applying it to the first tree and the second, whether this interval was a minute or an age, it is obvious that it  could not serve as a measure of their relative length: there would in that case be no common medium of comparison. It is essential to the discovery of the mutual relation of two objects, which cannot be directly compared, that their respective relations to some third object should be known: but in this case, the ratio which the trees were found to bear would not be to the same object, and therefore nothing could be told as to the ratio of the trees to each other. It is thus indispensable, that the instrument employed as a measure should remain unaltered, or be altered in a known degree, during its successive applications to the objects measured, in order to give us their relations to one common object. By this means we obtain a common term or denomination, in which the lengths of the two trees are expressed. This is, in fact, all that is essential to the end in view : the measurement, that is, the actual application of the physical instrument to the object, is the means, and the unvarying length of the instrument, or its ascer-tainable variation during the process, is the necessary condition for obtaining that common expression of the length of the two objects, which will show their relation to each other. But it is obvious that this relation of length would be equally determined in whatever way the common expression was obtained.
Now in the case of value, we obtain this common expression without that physical operation here described. We learn the values of two commodities in relation to the third, not from the application of an instrument, first to one commodity and then to the other, but from intercourse with mankind, or from the inspection of documents in which they are registered. We equally obtain a common expression, but we obtain it by different means. But the invariableness in the length of a measure of space, as above described, is a circumstance belonging to the means employed to obtain a common expression of length; and as the means of obtaining a common expression of value are totally different, as in fact the common expres-sion is necessarily implied in the supposition of using any commodity as a medium of comparison, there is nothing in the latter case in which invariableness of any kind, or in any sense, can be required. In the one case there is an instrument employed in a physical operation, and it is for the purpose of rendering this instrument capable of performing its function, that invariableness is indispensably necessary : in the other case there is no instrument so employed, and therefore there is no invariableness wanted: in the former case invariableness in the instrument (under the modification which it is needless to repeat) is essential to the attainment of the common term; in the latter, the common term being given, there is nothing in which invariableness can have place, or of which it can be predicated. If the length of the rod varied in an unknown degree between applying it to the two objects, we should have two terms of unknown relation to each other, and there could be no comparison of the objects to be measured ; and if the values of the  two commodities, which we were desirous of comparing, were expressed in different media, there would be the same impossibility. Hence, if in the case of value we were under the necessity of finding a counterpart to invariableness of length in the instrument employed to compare the dimensions of two objects, it would be, not invariableness of value in the commodity used as a medium to compare the value of two other commodities, but the condition that the value of these commodities should be given in relation to the same medium, or, in other words, expressed in a common denomination.
From all this it appears, that the analogy universally supposed to exist in this matter is altogether imaginary, and the phrase, invariable measure of value, proves to be absolutely destitute of a basis of meaning.
The doctrine which exacts invariableness in a measure of value, furnishes one corollary, which has been so frequently maintained and so generally adopted, that although its refutation is contained in the preceding observations,  it appears to require a separate examination. It is argued, that money or any other commodity is a good measure of the value of commodities, only at the same time, because it is liable to vary; while to perform this function correctly, there should be a commodity the value of which did not vary from one age to another; as to measure the lengths of objects at different periods, there must be an object of invariable length *. Let us therefore endeavour to ascertain what this really amounts to. With regard to the measurement of space, the intervention of time occasions no alteration in the requisite conditions. The preceding remarks are as applicable to the measurement of the length of objects at different times as to the same time. The qualification necessary to constitute an instrument a good measure of space, is in each  case identical, namely, invariableness during its application to all the objects compared. Whether an hour or a century elapses between the successive applications of the instrument makes no difference. The essential requisite is the same in measuring objects in distant ages, or objects existing at the same time.
But in the process called measuring value, there is no application of any instrument, and therefore, as I have already shown, there is absolutely nothing to which the quality of invariableness can be attributed, or of which it can be affirmed. The requisite condition in the process is, that the commodities to be measured should be reduced to a common denomination, which may be done at all times with equal facility; or rather it is ready done to our hands, since it is the prices of commodities which are recorded, or their relations in value to money. If money, therefore, is a good medium of comparison at one time, it is at all times.
 It may be objected, Yes, good enough for commodities at each time, but not between commodities at different times.
This objection, however, proceeds on a fundamental mistake already exposed in a former chapter, namely, that the relation of value can exist between commodities at different periods, which is in the nature of the case impossible; and if no relation exists there can be no measurement of it. It is, in truth, only the value of commodities at the same time that can be measured; another point in which the supposed analogy between the measurement of space and of value completely fails. In the case of length, a direct comparison may be made between two objects, however separated by time, and their ratio to each other found. The length of an object now may be compared with the length of an object in former times, by means of an instrument actually handed down to us ; by an uninterrupted transmission of the same object, or the same space through the medium of different objects, furnishing a com-mon bond of connection between the measurements of space in all ages. But this circumstance can evidently have no existence in the measurement of value, which is the ascertainment of a relation between contemporary commodities, and not between objects at different periods. The two cases would be analogous if we supposed no physical measure of length to be transmitted from one period to another, but only a record of the lengths of different objects expressed in a common denomination. Under these circumstances, all that we could do would be to compare the relative dimensions of objects in our own days, with the relative dimensions of similar objects in past times, as recorded: but we should have no common medium of comparison between one age and another. Now what in this case would be owing to the want of a transmitted measure, arises in the other case from the very nature of the relation with which we have to do. The nature of that relation itself interposes as complete a disconnection between different ages, as  would result from the supposed want of a common instrument for measuring space.
It is obvious then, that if no relation of value can exist between objects in different ages, there can be no measurement of it, nor consequently can there be any measure or medium of comparison required.
The only thing to be done, with regard to different periods, is to compare the relation of value subsisting between any two commodities, A and B, at one period, with the relation subsisting between them at another; or, in other words, the quantity of A which purchased B at the former time, with the quantity of A which purchased B at the latter. This is evidently a simple comparison, in which neither A nor B perform the function of a measure, or medium, in any possible interpretation of the term. That office has in all likelihood been already discharged in ascertaining the relative quantities of A and B at each period; and if, as is probable, these quantities have been ascertained by means of the prices of the commodities, money  has been the medium of comparison. But after these quantities have been ascertained, there can be no place whatever in the subsequent comparison for any medium, no conceivable function for it to perform.
Should it be urged, that when we compare the price of corn in one year with its price in another, we use money as a medium of comparison, in the same way as when we compare the prices of corn and cloth at the same point of time, the answer is not difficult.
In the latter case it is obvious, that the facts furnished to us are the relations of cloth and corn to money, or, the quantities of money for which definite portions of them are exchanged; and from these we infer another circumstance, namely, the relation of value between corn and cloth, and consequently their comparative power of purchasing all other commodities.
In the former case, on the other hand, the facts furnished to us are the prices of corn, or the relations between corn and money, at two different periods: but from these we deduce no other rela-tion; we do not advance a step beyond the information given; there is no inference corresponding to that which is drawn in the other case. We cannot deduce the relation of value, between corn at, the first and corn at the second period, because no such relation exists, nor consequently can we ascertain their comparative power over other commodities. If we made the attempt, it would be in fact endeavouring to infer the quantities of corn which exchanged for each other at two different points of time, a thing obviously absurd. And further, money would not be here discharging a particular function any more than the other commodity. We should have the value of corn in money, and the value of money in corn, but one would be no more a measure or medium of comparison than the other.
These observations are enough to show, that the only use of a measure of value, in the sense of a medium of comparison, is between commodities existing at the same time; and consequently the proposition, that money is not  a good measure of the value of' commodities at different periods, is either false or amounts to nothing. If it means that money is not equally a good measure of contemporary commodities at any period, it is directly opposite to the truth: if it means that it is not a good medium of comparison between commodities at different periods, it asserts its incapability of performing a function in a case where there is no function for it to perform.
In applying the principles developed in the preceding disquisition to the writings of Mr. Ricardo, we shall find that he has fallen into the same errors as his predecessors and contemporaries, as well as into others peculiarly his own. Misled by his radical misconception of the nature of value, and particularly by his notions on the subject of real value, he has opened his section on an invariable measure, with the following passage, the errors of which will be sufficiently apparent to any one who has attended to the foregoing part of the present chapter.
When commodities, says Mr. Ricardo,  varied in relative value, it would be desirable to have the means of ascertaining which of them fell, and which rose in real value, and this could be effected only by comparing them, one after another, with some invariable standard measure of value, which should itself be subject to none of the fluctuations to which other commodities are exposed. Of such a measure it is impossible to be possessed, because there is no commodity which is not itself exposed to the same variations as the things, the value of which is to be ascertained ; that is, there is none which is not subject to require more or less labour for its production.
It has been already shown in the first chapter, not merely that such a commodity is physically impossible, as here conceded by Mr. Ricardo, but that the supposition of such a commodity, for such a purpose, involves contradictory conditions*. We could not in the nature  of the case have any commodity of invariable value, by which to ascertain the fluctuations of all other things, unless all commodities were of invariable value, in which case there would be no fluctuations to ascertain.
We have also seen in the present chapter, that the demand for invariablenes of value in any commodity to be used as a measure, is founded altogether on a false analogy; that fluctuations in value are not ascertained by any measure, but by historical evidence; that a measure of value can signify nothing but a medium of comparison for contemporary commodities; and that we have as good a measure in this sense, not only as it is possible to have, but as it is possible to conceive.
Besides these errors, there is to be discovered in Mr. Ricardo's views, as to the uses of a measure of value, a singular confusion of thought, which I shall here endeavour to explain.
The specific error of Mr. Ricardo on the subject of invariable value consists, as before explained, in supposing, that if the causes of  value affecting one commodity remained the same, the value of that commodity could not vary, overlooking the circumstance, that value denotes a relation between two objects, which must necessarily alter with an alteration in the causes affecting either of them. He incessantly identifies constancy in the quantity of producing labour with constancy of value, Hence he maintains, that if we could find any commodity invariable in the circumstances of its production, it would be in the first place invariable in value; and, secondly, it would indicate, or would enable us to ascertain, the variations in value of other commodities.
It is curious enough that he should never have clearly discerned what such a commodity would really serve to indicate: it would not, as he asserts, serve to indicate the variations in the value of commodities, but the variations in the circumstances of their production. It would enable us to ascertain, not any fluctuations in value, but in which commodity those fluctuations had originated. He has in truth  confounded two perfectly distinct ideas, namely, measuring the value of commodities, and ascertaining in which commodity, and in what degree, the causes of value have varied.
For suppose we had such a commodity as he requires for a standard : suppose, for instance, all commodities to be produced by labour alone, and silver to be produced by an invariable quantity of labour. In this case silver would be, according to Mr. Ricardo, a perfect measure of value. But in what sense? What is the function performed ? Silver, even if invariable in its producing labour, will tell us nothing of the value of other commodities. Their relations in value to silver, or their prices,. must be ascertained in the usual way, and when ascertained, we shall certainly know the values of commodities in relation to each other : but in all this there is no assistance derived from the circumstance of the producing labour of silver being a constant quantity.
But it is the fluctuations of commodities which this invariable standard is to ascertain or  measure. Let us try to discover how far it would assist us here.
Suppose cloth in the year 1600 was worth 12s. a yard, and in 1800 only 6s., Here we have a fluctuation in the value of cloth, in relation to the standard commodity; in 1800 it was worth only half as much silver as it was in 1600. This, however, is not, let it be observed, a fluctuation ascertained by the circumstance of silver being produced by an invariable quantity of labour. Had silver varied in the circumstances of its production, our information as to the relation between cloth and silver would have been equally attainable, and equally complete. What then could be ascertained, in this case, from the metal being invariable in the quantity of its producing labour? What inference would this circumstance enable us to draw ? No inference, obviously, as to the value of cloth and silver; for, on this point, the prices of the former tell us all that it is possible to know. The inference we should draw would be, that the  cause of the chance in the relation between cloth and silver had been in the former, and as labour is, by the supposition, the sole cause of value, we might more particularly infer, that the producing labour of cloth had been abridged to half its former quantity.
A commodity, therefore, under these conditions, produced by an invariable quantity of labour, would enable us to ascertain, not the fluctuations in value between two or more commodities (for these are facts to be gathered from appropriate evidence), but the fluctuations in the quantity of labour which produced them: and in truth,. if we examine what is the particular advantage which Mr. Ricardo himself supposes we should be able to derive from the possession of such a commodity, we shall find it to be in reality that which is here described, the power of ascertaining, not the variations in value, but the variations in the producing labour of commodities. Speaking of the interchange of game and fish, in the earlier stages of society, he says,
 If with the same quantity of labour, a less quantity of fish, or a greater quantity of game were obtained, the value of fish would rise in comparison with that of game. If, on the contrary, with the same quantity of labour a less quantity of game, or a greater quantity of fish was obtained, game would rise in comparison with fish.
If there were any other commodity, which was invariable in its value, we should be able to ascertain, by comparing the value of fish and game with this commodity, how much of the variation was to be attributed to a cause which affected the value of fish, and how much to a cause which affected the value of game.
Suppose money to be that commodity. If a salmon were worth £1, and a deer £2, one deer would be worth two salmon. But a deer might become of the value of three salmon, for more labour might be required to obtain the deer, or less to get the salmon; or both these causes might operate at the same time. If we had this invariable standard, we might easily as  certain in what degree either of these causes operated. If salmon continued to sell for £1, whilst deer rose to £3, we might conclude that more labour was required to obtain the deer. If deer continued at the same price of £2, and salmon sold for 13s. 4d. we might then be sure that less labour was required to obtain the salmon; and if deer rose to £2. 10s. and salmon fell to 16s. 8d. we should be convinced that both causes had operated in producing the alteration of the relative value of these commodities.
Here we have a very accurate description, by Mr. Ricardo, of what a commodity produced by an invariable quantity of labour (not a commodity of invariable value, as he erroneously terms it) would enable us to ascertain, under the supposition that all things were determined in value by quantity of labour. He does not tell us that such a commodity would enable us to ascertain the value of fish or game, or their variation in value, but this variation being given, that it would enable us to infer how much of it was to be attributed to a  change in the labour required to obtain the salmon, and how much to a change in that required to obtain the deer.
In this and other passages it will be found, that although Mr. Ricardo is professedly speaking of a commodity produced by invariable labour, in the character of a measure of value, he is in reality, without being conscious of the difference, altogether occupied with the consideration of that commodity as capable of indicating variations in the producing labour of other commodities*. Instead of a measure of value,  such a commodity as he describes would be a measure of labour, or a medium of ascertaining the varying quantities of labour which commodities required to produce them. Before it could be employed in regard to any object, the value of that object, or its relation to the standard commodity, must be given, and then all that could be deduced from the datum would be the quantity of labour bestowed on its production.
But perhaps the most remarkable circumstance of all is, that for this latter purpose, that invariableness in the quantity of labour, which he has insisted upon as so essentially requisite, would be of no peculiar service. On the supposition that labour was the sole determining principle of value, a commodity produced by an invariable quantity of labour would afford us no assistance even as a measure of labour, which could not be equally derived from a commodity the producing labour of which was variable, provided we were furnished with the same data.
 For in the above comparison of cloth in 1600 and cloth in 1800, mark all that is specifically ascertained.
If silver had been liable to variation in the quantity of its producing labour, we should still have been informed, from the same source that supplied the information in the other case, what was its relation to cloth, for this is equivalent to saying, that we should still have been informed of the prices of cloth at the two different periods specified. These axe historical facts, and not deductions from the invariableness of the labour employed in the production of silver. Were this labour then a variable quantity, we should still learn, that a yard of cloth in 1600 was 12s. and in 1800 6s. ; but we should, it is alleged, be at a loss to discover, whether the change in the relation between silver and cloth had been owing to the former or the latter. This then is the sole circumstance by which the two cases are supposed to be distinguished, and in fact it amounts to this; we could tell that, in the former case, cloth in 1800 required only  half the labour necessary for its production in 1600, while in the latter case we could not tell whether the quantity of producing labour in the cloth had been reduced one half, or whether that required for the production of money had been doubled. In answer to this I say, that the ratio between the quantities of labour necessary for the production of cloth in 1800, and in 1600, might be equally ascertained, although the quantity of labour employed in the production of silver had varied, provided that the data in the two cases were equal.
The data in the first case are the prices of cloth at each period, and the ratio subsisting between the; quantity of labour employed at each period in the production of silver. The circumstance of this ratio being that of equality makes no difference.
Now suppose, in the second case, that we are furnished with the prices of cloth at both periods, and with the ratio subsisting between the quantities of the labour necessary for the production of silver, which ratio, by the sup-position, not being that of equality, suppose to be as 2 in 1600 to 1 in 1800, or, in other words, suppose that silver in 1800 is produced by half the labour required in 1600.
With these data it is obvious, that we could deduce the ratio of labour employed in the production of cloth at these periods, with as much accuracy as we could under the conditions of the first case. If in 1600 the cloth was 12s. per yard, and in 1800 only 6s., the producing labour of silver at the latter period being only half of what it was at the former period, then the producing labour of cloth would have been reduced to a quarter of its former quantity. For in 1600 a yard of cloth being 12s. in value, the yard of cloth and the 12s. took equal quantities of labour to produce them: but in 1800 the producing labour of 12s. is by the supposition reduced one half, and consequently the quantity of labour in 6s. must be a quarter of the quantity which had been necessary to produce 12s. in 1600. Now as 6s. in 1800 exchange for a yard of cloth, the pro-ducing labour of the yard of cloth must be equal to the producing labour of the 6s.; that is, a quarter of the quantity of labour employed to produce a yard of cloth in 1600.
It may probably be alleged, however, as an advantage peculiar to the first case, that the quantity of producing labour being invariable, we are saved from all that research into its comparative quantity at different periods which would be necessary on the contrary supposition. But it is to be recollected, that the circumstance of a commodity having been always produced by the same quantity of labour, is an historical fact quite as difficult to ascertain as the variations of another commodity. We might, it is true, be saved from all investigation of this nature, if there existed a commodity, which, from some obvious and insuperable necessity, was always the product of the same labour; yet even this advantage is not dependent on the invariableness of the labour; for if, what is equally easy to suppose, and quite as likely to happen, we had a commodity which necessarily  varied every year in a given proportion, we should be equally spared the pains of historical research. To have a commodity, whether produced by a variable or by an invariable quantity of labour, which saved us the trouble of inquiry, would doubtless be an advantage, but we might as well suppose fifty other arbitrary aids *.
In concluding this discussion, it may not be useless to advert more particularly to one of the objects, which economists have proposed to themselves in the attempt to discover an invariable measure or standard of value. It appears to have been to determine the efficiency of revenues, salaries, and wages of different classes of people at different periods, in what condition such revenues enabled them to live, or what power it enabled them to wield. This, it is supposed, would be accomplished, did we possess some object of immutable value.
If we are told, says Mr. Malthus, that the wages of day-labour in a particular coun-try are, at the present time, four pence a day or, that the revenue of a particular sovereign, 700 or 800 years ago, was 400,000l. a year, these statements of nominal value convey no sort of information respecting the condition of the lower classes of people in the one case, or the resources of the sovereign in the other. Without further knowledge on the subject, we should be quite at a loss to say, whether the labourers in the country mentioned were starving, or living in greater plenty; whether the king in question might be considered as having a very inadequate revenue, or whether the sum mentioned was so great as to be incredible.
It is quite obvious, that in cases of this kind, and they are of constant recurrence, the value of wages, incomes, or commodities, estimated in the precious metals, will be of little use to us alone. What we want further is some estimate of a kind which may be denominated real value in exchange, implying the quantity of the necessaries and conveniences of life, which those wages, incomes, or commodi-dities will enable the possessor of them to command *.
Now to suppose that we can have any one object by which this information can be obtained, would imply a gross misconception of the nature of value. I have already repeatedly stated, that to know the value of an article at any period, is merely to know its relation in exchange to some other commodity. From this fact, which must be ascertained like other facts, no inference whatever can be drawn as to the value of any thing beyond the two commodities in question. From the relation of corn and money nothing can be inferred as to the relation of corn and labour, or of money and labour. If, proceeding a step farther, we learn from the proper records the relation also of labour and money, then we can deduce the relation of labour and corn; but we should not be able to make any inference to any other object. The only practicable Inference on the subject  of value, is the mutual relation of two commodities from their separate relations to a third.
It follows, that if we wish to ascertain the state of comfort or luxury in which any class of people lived at any assigned period, there is no possible method of effecting the object, but ascertaining from the proper documents the amount of their incomes, and then, particular by particular, the relation which these incomes bore to commodities. If the incomes are stated in corn, or silver, nothing can be inferred from the statement, as to their power over other things. Supposing the income to be a certain amount of money, then the inquirer must find records of the prices of those articles to which his curiosity is directed, and a simple calculation will teach him the power of the income to command them.
If he wishes, for example, to ascertain the condition of the labouring class at any given period, he must first find the rate of wages, or, in other words, the mutual relation of labour and money. This is one step in the investiga-tion, but it will not of itself throw any light on the food, clothing, and comfort, which the labourers are able to procure; and he must therefore search in the proper registers for the prices of such commodities as constitute these necessaries and conveniences. He can ascertain nothing but what is shown by the historical documents which he consults. When he has found the price of labour, the price of corn, of cloth, of hats, of stockings, of fuel, of house-room, he will be able to tell how much of each of these commodities a week's or a year's labour could command : in other words, the condition of the labouring class of society in these respects will become manifest.
But these are all separate particulars, to be separately ascertained: one will not disclose another; each must be individually established by independent evidence. There can be no commodity, by a reference to which the power of a given income over any or all other commodities may be shown.
 Conclusions such as these are so obvious, that they would scarcely require to be formally stated had they not been frequently overlooked. Even the author of the Templars' Dialogues, who observes, that Mr. Malthus, in common with many others, attaches a most unreasonable importance to the discovery of a measure of value, seems to sanction the prevailing errors, when he goes on to remark, that such a measure would at best end in answering a few questions of unprofitable curiosity*.  Sufficient, it is hoped, has been said to show, that we are in possession of the only kind of measure which can be had or conceived, and that we must look for the gratification of our curiosity, not to any measure of value whatever, but to the records of former times, and a few simple calculations from the data which they furnish.