P.N. Bhagwati, C.J.
1. This reference arises out of an assessment to income-tax made of the assessee for the assessment year 1964-65, the corresponding account year being the financial year ended 31st March, 1964. The assessee had a large area of ancestral agricultural land admeasuring 2,02,088 square feet situate in village Savad at some distance from Baroda City. Two agreements of sale were entered into by the assessee in respect of different portions of this land, one on 27th June, 1962, and the other on 7th July, 1962. By the agreement of sale dated 27th June, 1962, the assessee agreed to sell 2 acres 2 gunthas of land was agreed to be Company of Baroda and the remaining 30 gunthas of land was agreed to be sold to Bapunagar Co-operative Housing Society Ltd., by the agreement of sale dated 7th July, 1962. The land was being used for agricultural purposes by the assessee and it was, therefore, apparent that by reason of section 63 of the Bombay Tenancy and Agricultural Lands Act, 1948 (herein-after referred to as 'the Tenancy Act'), the assessee would not be able to sell any portion of the land to Ajay Construction Company or Bapunagar Co-operative Society Ltd., pursuant to the agreements of sale unless the land ceased to fall within the definition of 'land' in section 2(8) of the Tenancy Act. The assessee accordingly applied to the Collector under section 65 of the Bombay Land Revenue Code, 1979, for permission to make non-agricultural use of the land and such permission was granted on 23rd January, 1963. The assessee admittedly ceased to carry on agricultural operations on the land from 23rd January, 1963, and completed the sales in favour of Ajay Construction Company and Bapunagar Co-operative Housing Society Ltd., on 29th April, 1963, respectively for the sums of Rs. 30,000 and Rs. 50,000. In the course of assessment of the assessee to income-tax for the assessment year 1964-65, the Income-tax Officer took the view that at the dates when the sales were effected, the portions of land sold had ceased to bear the character of agricultural land and were 'capital assets' within the meaning of section 2(14) of the Income-tax Act, 1961, and capital gain arising from the sales of these capital assets was, therefore, liable to be included in the taxable income of the assessee. The question then arose as to how the capital gain should be computed. According to section 48 'the cost of acquisition of the capital asset' is to be deducted from the amount of consideration for determining the capital gain. The assessee contended that the land was throughout agricultural land and it became 'capital asset' in the hands of the assessee for the first time on 23rd January, 1963, when it was converted into non-agricultural land and, therefore, 'the cost of acquisition of the capital asset' in relation to the must be taken to be the market value of the land on 23rd January, 1963. This contention was negatived by the Income-tax Officer who held that 'the cost of acquisition of the capital asset' was what if cost the assessee to acquire the capital asset, namely, the land, and since the land was acquired by the assessee prior to 1st January, 1954, the assessee had the option under section 55(2) to substitute the fair market value of the land on 1st January, 1954. The Income-tax Officer accordingly computed the capital gain by deducting from the sale proceeds the market value of the land on 1st January, 1954. There was a claim for deduction of Rs. 10,000 made by the assessee in respect of the amount paid to the tenant for surrender of the tenancy rights but this claim was disallowed by the Income-tax Officer. The result was that a sum of Rs. 40,000 was brought to tax as capital gain in the hands of the assessee. The assessee appealed to the Appellate Assistant Commissioner but the appeal was unsuccessful and hence the matter was carried in further appeal to the Tribunal. The Tribunal rejected both the contentions of the assessee in regard to the taxability of the amount but allowed the deduction of Rs. 10,000 in computing the capital gain and reduced the amount of capital gain to Rs. 30,000. The assessee was obviously dissatisfied and he, therefore, moved for a reference and, on his application, the following two questions of law, namely :
'(1) Whether, on the facts and in the circumstances of the case, the lands sold by the assessee were agricultural lands within the meaning of section 2(14) of the Income-tax Act, 1961 and
(2) If the answer to question No. (1) is in the negative, whether, in computing the capital gains, the assessee is entitled to deduct from the sale market value as on January 23, 1963, when the lands were converted into non-agricultural lands ?'
were referred by the Tribunal for the opinion of this court. The first question is whether the land sold by the assessee was 'agricultural land' within the meaning of section 2(14) at the date when under section 45 the charge to income-tax under the head 'capital gains' is attracted only in respect of profits or gains arising form the transfer of a 'capital asset' and 'capital asset' is defined in section 2(14) to mean property of any kind held by an assessee but, including, inter alia, not agricultural land in India. If, therefore, the land sold by the assessee was agricultural land at the dates when the sales of its different portions were effected, profits or gains arising from such sales would not be taxable as capital gain under section 45. Now, the expression 'agricultural land' is nowhere defined in the Act and it must, therefore, be understood in its plain ordinary meaning according to English language. What is it that according to the ordinary natural sense of the expression distinguishes 'agricultural land' from other land We had occasion to consider this question in a recent decision given in a Wealth-tax Reference, Rasiklal Chimanlal Nagri v. Commissioner of Wealth-tax.
2. There the question was whether certain land belonging to the assessee was agricultural land within the meaning of section 2(a) of the Wealth-tax Act so as to be exempt from wealth-tax and that necessitated consideration of the true meaning and import of the expression 'agricultural land'. Though the observations made in that decision were in reference to the expression 'agricultural land' as occurring in section 2(e) of the Wealth-tax Act, what we said there must apply equally in determining what is 'agricultural land' within the meaning of section 2(14) of the Income-tax Act. We pointed out various factors which have to be taken into account in determining whether a particular land is agricultural land and this is what we said :
'One thing is clear that the intention of the owner of the land to put it to any particular use at a given point of time cannot be the determining factor. Whether a land is agricultural land or not cannot depend on the fluctuating or ambulatory intention of the owner of the land. The criterion must be something more definite, something more objective, something related to the nature or character of the land and not varying with the intention of the owner as to the use to which he wants to put the land at a particular point of time. Of course, when we say this we must not be understood to mean that the intention as to user is altogether an irrelevant consideration; it is certainly a factor which would bear on the nature or character of the land but it does not afford a sole or exclusive criterion for determined whether a land is agricultural land or not. Where the land is actually put to use, there is usually not much difficulty in ascertaining the nature or character of the land. If the land is used for agricultural purposes, ordinarily it would be correct to say that the land is agricultural land and vice versa. But even this test may not always furnish a correct answer, for there may be cases where land admittedly non-agricultural (such as a building site) may be used temporarily for agricultural purposes. In such cases it would not be correct to say that merely because the land is in fact being used for agricultural purposes, it is agricultural land. But as a general proposition it may be stated without any fear of contradiction that ordinarily the actual user to which the land is being put would furnish, prima facie, evidence of the true nature or character of the land and, therefore whenever a question arises whether a particular land is agricultural land or not, primarily regard must be had to the purpose for which the land is being actually used at or about the relevant time and that would ordinarily provide a satisfactory answer to the problem.......... Whether a particular land is agricultural land or not must depend on the general nature or character of the land, and various factors would have to be taken into account. The development and use of the lands in the adjoining area and the surroundings and situation of the land would be an important factor which would have a bearing on the question whether the land is agricultural land or not. The factor may affect the land and its capacity of being used for agriculture and would also indicate the purpose for which the land would ordinarily be likely to be used. The physical characteristics of the land would be another factor to be taken into account..... Then the intention of the owner as gathered from all the relevant circumstances would also have a bearing on the general nature or character of the land to put it to a particular use at any given point of time cannot be the determining factor. But the intention of the owner in regard to the user of the land would certainly be a relevant factor which would have to be taken into account. Where, for example, as in the present case, the land has not been used for agricultural purpose for over a number of years without any particular reason, it would certainly indicate that the land is no longer meant for agricultural purposes, but is meant for being used for non-agricultural purposes and cannot, therefore, be regarded as agricultural land. The fact that the land is assessed for agricultural purpose would also be a relevant consideration and due effect would have to be give to this factor in arriving at the conclusion whether the land is agricultural land or not. But we cannot agree that the capacity of the land for being put to agricultural use is a determinative factor in deciding whether the land is agricultural land or not. If that were the correct test, even building sites assessed for non-agricultural purpose would be agricultural lands so long as they are not actually put to non-agricultural use, since it would always be possible to say to them that they are capable of being used for agricultural purpose. As a matter of fact all land which has not actually been put to non-agricultural use would be liable to be regards as agricultural land if this test were the correct test.'
3. It will be clear from these observations that the true test to be applied for the purpose of determining whether a particular land is agricultural land or not is first to ascertain what is the use to which the land is being actually put. If it is being used for agricultural purpose or even if the agricultural use has ceased but it is apparent that the land is meant to be used for agricultural purpose, it would be agricultural land. If on the other hand the land is being used for non-agricultural purpose, it would be a strong circumstance to indicate that the land is not agricultural land. Where, however, the land is not being actually put to any use, the test would be not whether the land is capable of being used for agricultural purpose but whether, having regard to the various factors referred to in the above-quoted passages from the judgment in Rasiklal Chimanlal Nagri's case, the general nature or character of the land is such that it can be regarded as agricultural land. If we examine the question from this point of view, the conclusion is inescapable that the land sold by the assessee in the present case was not agricultural land at any time when one or the other portion of it was sold. It is no doubt true that prior to 23rd January, 1963, the land was used for agricultural purpose by the assessee but the agricultural user came to an end on 23rd January, 1963, when permission for non-agricultural use was obtained from the Collector. This cesser of agricultural use was not of a temporary nature leaving the agricultural character of the land unaffected. There are various circumstances which indicate plainly and unmistakably that the land was at no time thereafter intended to be used for agricultural purpose. In the first place, the permission for non-agricultural use obtained from the Collector clearly evinced the intention of the assessee that the land would thereafter be used for non-agricultural purpose. Secondly, it was common ground that the land was agreed to be sold to Ajay Construction Company, Baroda, and Bapunagar Co-operative Housing Society Ltd. for construction of residential house and the user of the land was, therefore, intended to be permanently changed : the land was no longer to be used for agricultural purpose but it was to be used for construction of residential houses. The general natures and character of the land was thus altered and, from agricultural land, it became non-agricultural land. There is also one other circumstance which strongly reinforces this conclusion. If the land continued to be agricultural land, the assessee could not have sold it to Ajay Construction Company, Baroda, and Bapunagar Co-operative Housing Society Ltd. in view of the provisions of section 63 of the Tenancy Act and the assessee, therefore, with a view to escaping from the mischief of section 63 of the Tenancy Act, obtained permission of the Collector for non-agricultural use under section 65 of the Bombay Land Revenue Code, 1879. The object of the assessee was that the land should cease to fall within the definition of 'land' in section 2(8) of the Tenancy Act and, if that was the object, it must follow necessarily that the general nature or character of the land was converted from agricultural into non-agricultural. It may also be noted that the price at which the land was sold by the assessee was fixed not by acre, vigha or guntha but by square foot. It is common knowledge that when agricultural land is sold, the price is almost always fixed with reference to acre, guntha or vigha. The fixation of price with reference to square food it generally a characteristic of non-agricultural land. These circumstances clearly establish that, though originally the land sold by the assessee was agricultural land, it ceased to be agricultural land and became non-agricultural land from 23rd January, 1963, and, therefore, at the time of sale, it was not agricultural land within the meaning of section 2(14) of the Act. The first question must accordingly be answered in the negative.
4. That takes us to the second question which raises a controversy as to how capital gain is to be computed when a capital asset which is sold was not 'capital asset' at the date when it was acquired by the assessee but subsequently became 'capital asset', as for example, in the present case, where the land being originally agricultural was not capital asset but acquired the character of capital asset on 23rd January, 1963, when it ceased to be agricultural land and became non-agricultural. To determine this controversy, it is necessary to refer to a few relevant provisions of the Act. Section 45 charges to tax 'any profits or gains arising from the transfer of a capital asset effected in the previous year' and provides that such profits or gains shall be deemed to be the income of the provides that such profits or gains shall be deemed to be the income of the previous year in which the transfer took place. It is clear on a plain reading of this section that the only condition for attracting the charge of tax under it is that the capital asset must be transferred : in other words, there must be transfer of property and the property transferred must be 'capital asset' at the date transfer are taxable as 'capital gains'. So far as the present case is concerned, as we have already pointed out above, the land sold by the assessee was not agricultural land at the time of sale and was, thereof, clearly 'capital asset' and its transfer by way of sale attracted the charge of tax under section 45. But the question is : how are the profits or gains arising from the transfer to be computed The answer is provided by section 48, which says :
'The income chargeable under the head 'capital gains' shall be computed by deducting from the full value of the consideration received or accruing as a result of the transfer of the capital asset, the following amounts, namely :-
(i) expenditure incurred wholly and exclusively in connection with such transfer; and
(ii) the cost of acquisition of the capital asset and the cost of any improvement thereto.'
5. To determine the profits or gains arising from the transfer of a capital asset, section 48 requires that there shall be deducted from the full value of the consideration received or accruing as a result of the transfer of the capital asset, 'the cost of acquisition of the capital asset.' Section 55, sub-section (2) defines what shall be the meaning of the expression 'cost of acquisition' in relation to a 'capital asset' for the purpose of section 48. It says, omitting portions immaterial :
'55. (2) For the purpose of sections 48 and 49, 'cost of acquisition', in relation to a capital asset, -
(i) where the capital asset became the property of the assessee before the 1st day of January, 1954, means the cost of acquisition of the asset to the assessee or the fair market value of the asset on the 1st day of January, 1954, at the option of the assessee....'
6. Now the question which arises for consideration is as to what is the true meaning and import of the words' the cost of acquisition of the capital asset' in section 48, clause (ii). One construction suggested on behalf of the assessee was that these words, on a plain grammatical construction, require that the property which is sold must be 'capital asset' at the date of acquisition by the assessee, for, otherwise, these words would be rendered inappropriate and meaningless. But, this construction cannot be accepted and for two very good reasons. One is that it would introduce an additional condition of attracting the charge to tax which is not to be found in section 45 which is the charging section. The only condition for attracting the charge to tax which is laid down in section 45 is that the property transferred must be capital asset at the date of transfer. How the profits or gains arising from the transfer of such property are to be computed is laid down in section 48. Section 48 is not intended to lay down any further condition for attracting the charge to tax. It would not, therefore, be right to construe section 48, clause (ii), as providing that the property, besides being capital asset at the date of transfer as required by section 45, must also satisfy the definition of 'capital asset' at the date of acquisition by the assessee. Moreover, such a construction would stultify the charging provision by unduly restricting the ambit and scope of the both terminals, namely, date of the acquisition as in the present case, profits or gains arising from transfer of such property would be left out from the ambit and coverage of the charging provision and would escape tax. Of course, if such a result is inevitable, the court would not strain the language of the statutory provision in order to bring such cases within the net of taxation but if, on a plain natural construction of the language used by the legislature, it appears that such cases were also intended to be covered by the statutory provision and there is moreover no rational justification for leaving out such cases from the scope and ambit of the charging provision, we would not be reluctant to adopt a construction which would bring such cases within the charging provision. It is, therefore, apparent that the only condition which must be satisfied in order to attract the charge to take under section 45 is that the property transferred must be capital asset at the date of transfer and it is not necessary that it should have been capital asset also at the date of acquisition by the assessee.
7. But then what meaning is to be given to the words 'the cost of acquisition of the capital asset' How is full effect to be given to these words if the property transferred need not be capital asset at the date of acquisition The answer to this question is simple if only we read these words in the light of section 45. Section 45 says that profits or gains arising from the transfer of a capital asset shall be chargeable to take and section 48, clause (ii), then proceeds to add that such profits or gains shall be computed by deducting from the full value of the consideration for the transfer 'the cost of acquisition of the capital asset.' The words 'the capital asset' in section 48, clause (ii), are clearly intended to refer to the capital asset which is transferred as mention in section 45. They are identificatory words to denote the property transferred and they do not introduce any requirement that the property transferred shall be capital asset at the date of acquisition. The law says that when property which is a capital asset is transferred, profits or gains arising from the transfer shall be liable to tax and you shall compute such profits or gains by deducting from the consideration for the transfer, what cost you to acquire the 'capital asset', that is, the property transferred. The difference between the consideration for the transfer of the property and the cost of acquisition of the property would represent the profits or gains arising from under section 45. This appears to be the plain natural construction of the words used in section 48, clause (ii), read with section 45. There are also certain inherent indications in the Act which go to show that this is the correct construction. Section 55(2), clause (i), says that where the capital asset became the property of the assessee before the 1st day of January, 1954, the cost of acquisition in relation to the capital asset means the cost of acquisition of the asset to the assessee or the fair market value of the words 'where the capital asset became the property of the assessee' clearly show that the expression 'capital asset' is used as a demonstrative noun to refer to the property transferred. The question which has to be asked for the purpose of section 55(2), clause (i), is when did the property designated as a capital asset in section 45 become the property of the assessee. If it became the property of the assessee before 1st January, 1954, the assessee would have the option either to take the cost of acquisition of the asset to him or the fair market value of the property on 1st January, 1954, as the cost of acquisition for the purpose of section 48, clause (ii).
8. The argument of the assessee was that the words 'capital asset' in section 48, clause (ii), must have the meaning given in the definition of 'capital asset' in section 2(14) and, therefore, section 48, clause (ii), must be construed as referring to the cost of acquisition of the property as a capital asset. If the property was not capital asset at the date of acquisition but subsequently acquired the character of capital asset, it must be with reference to the date of the property becoming capital asset, that 'the cost of acquisition of the capital asset' must be determined. What is required to be deducted, said the assessee, is not the cost of acquisition of the non-capital asset but the cost of acquisition of the capital asset and, therefore, the crucial date must be date on which the property became 'capital asset'. It was on the basis of this contention that the assessee urged that 'the cost of acquisition of the capital asset' in the present case must be taken to be the market value of the land on 23rd January, 1963, and not the original cost of acquisition of the land. But this contention is plainly contrary to the language of section 48, clause (ii). It is difficult to see how this construction accords with the words 'the cost of acquisition of the capital asset'. These words emphasise two aspects : one is 'acquisition' and the other is 'cost'. The reference clearly is to the point of time when the capital asset is acquired and the cost of such acquisition is required to be deducted from the full value of the consideration. Where the property transferred was not capital asset at the date of acquisition but subsequently became capital asset was acquired by the assessee when it was converted to into a capital asset and how would it would be possible in such a case to determine the cost of acquisition. There are no two different acquisitions of property, one as a non-capital asset and the other as a capital asset. The property is acquired by the assessee only once and merely its character changes in the sense that, whereas, originally it was non-capital asset, it now becomes capital asset. It would indeed be doing violence to the language of section 48, clause (ii), to read the words 'the cost of acquisition of the capital asset' in the manner suggested on behalf of the assessee. We would have to introduce an unwarranted fiction, namely, that when the property, which at the date of acquisition was non-capital asset, becomes capital asset, it is deemed to be acquired by the assessee as a capital asset on that date and, furthermore, though there can be no cost of such acquisition, the market value of the property on that date should be deemed to be the cost of such acquisition. There is no warrant for imposing such legal fiction on the plain language of section 48, clause (ii). The only justification which could be put forward on behalf of the the assessee for reading the section in this manner was that the legislature could not have intended that the appreciation in value which took place whilst the property was non-capital asset should be subjected to tax which would be the inevitable result if we read the section as referring to the cost to which the assessee was put in acquiring the property. But this is a wholly erroneous approach in construing a statutory provision. The intention of the legislature must be gathered from the words used; it is well settled that what is unexpressed by the legislature must be taken as unintended. We cannot presume a certain intention on the part of the legislature and then bend the language of the section with a view to making it accord with such presumed intention. The contention of the assessee also stands refuted by the language of section 55(2), clause (i). The property which is transferred could become the property of the assessee only at one point of time. It could not become the property of the another point of time. The argument of the assessee would require us to assume that when property which was non-capital asset becomes capital asset, it is deemed to become the property of the assessee for the purpose of section 55(2), clause (i). Such a construction would do violence to the language of section 55(2), clause (i), and would be clearly impermissible on the on sections 49, 51 and 55(3) that the words 'the cost of acquisition of the capital asset', 'the cost for which the asset was acquired' and 'the cost for which the previous owner of the property acquired it' are variously used by the legislature to denote the same idea and the reference is intended to be made only to the cost of acquisition of the property regardless of the question whether it was capital asset or non-capital asset at the date of acquisition.
9. It is, therefore, clear that the words 'the capital asset' in section 48, clause (ii), are identificatory and demonstrative and they are intended to refer the property which is the subject of levy of charge of capital gains under section 45. The Tribunal was, therefore, right in taking the view that, in computing profits or gains arising from the transfer of the land by at the option of the assessee, the fair market value of the land on 1st January, 1954, was liable to be deducted from the full value of the consideration under section 48, clause (ii), read with the section 55(2), clause (i). The second question referred to us for our opinion must be answered by saying that the assessee is entitled to deduct from the sale proceeds the market value of the land sold as on 1st January, 1954.
10. The assessee will pay the costs of the reference to the Commissioner.