South Africa: Supreme Court of Appeal

You are here:  SAFLII >> Databases >> South Africa: Supreme Court of Appeal >> 2012 >> [2012] ZASCA 88

| Noteup | LawCite

Distell Ltd v Commissioner for the South African Revenue Services (526/2011) [2012] ZASCA 88; 2012 (5) SA 450 (SCA) (31 May 2012)

Download original files

PDF format

RTF format

Links to summary

PDF format

RTF format

Bookmark/share this page

Bookmark and Share



SCA CASE NO: 526/2011


In the matter between:

DISTELL LIMITED …..............................................................................APPELLANT



AFRICAN REVENUE SERVICE ….....................................................RESPONDENT

Neutral Citation: Distell Limited v The Commissioner for the South African Revenue Service (526/2011)

[2012] ZASCA 88 (31 May 2012)

Coram: Navsa, Heher and Van Heerden JJA

Heard: 22 May 2012

Delivered: 31 May 2012

Summary: Excise and customs duty – Custom and Excise Act 91 of 1964 -classification of beverages under tariff headings – fermented or distilled (spirituous) beverages - International Convention on the Harmonised Commodity Description and Coding System.




On appeal from: North Gauteng High Court, Pretoria (Pretorius J sitting as court of first instance):

The appeal is dismissed with costs, including the costs of two counsel.




Navsa and Van Heerden JJA (HEHER JA concurring)

[1] This appeal involves a dispute about a tariff classification in relation to excisable goods under the Customs and Excise Act 91 of 1964 (the Act)1. The appeal turns on whether the products in question are fermented or distilled (spirituous) beverages. The appellants contended that they are fermented, and accordingly classifiable under a specific tariff heading, namely 22.05, alternatively 22.06, of part 1 of Schedule 1 to the Act. The respondent contended that they are spirituous, and therefore classifiable under tariff heading 22.08. Once that issue is determined the proper tariff item in part 2A of Schedule 1 under which the products should be classified will follow as a matter of course. Each tariff heading has a corresponding tariff item number. For ease of reference we shall refer only to the relevant tariff heading.

[2] The appellant company is Distell Limited (Distell), which owns and operates a number of wineries and conducts business as a manufacturer and distributor of liquor products. It markets and sells a number of well-known alcoholic beverages to commercial outlets. The respondent is the Commissioner for the South African Revenue Service (the Commissioner).

[3] The facts giving rise to the appeal are set out hereafter. During 2007


1 The products in question are goods manufactured in a customs and excise warehouse which renders them liable for the payment of excise duty: see s 37(1) of the Act.

and 2008 the Commissioner determined all of the products forming the subject matter of this appeal as falling within tariff heading 22.08 in Part 1 of Schedule 1 to the Act.2 The tariff headings themselves and an explanation of how they operate and are applied will be dealt with in due course. The products in question are the following:

(i) Angels' Share Cream;

(ii) Delgado Supremo;

(iii) GoldCup Creamy Vanilla;

(iv) Barbosa;

(v) GoldCup Banana Toffee;

(vi) Zorba;

(vii) Nachtmusik;

(viii) Mokador;

(ix) Alaska Peppermint;

(x) Copperband;

(xi) VinCoco;

(xii) Clubman Mint Punch;

(xiii) Viking;

(xiv) Castle Brand; and

(xv) Brandyale.

[4] As stated above, the tariff determinations were arrived at on the basis that the products in question are spirituous beverages. The Commissioner’s perspective, put simply, is that the base wines used in the beverages in question are subjected to processes in terms of which they are stripped of flavour and colour and have cane spirits added to them in order to bolster the alcohol content significantly, as well as sweeteners, flavourants and _____________________

2 Section 47(9)(a)(i) provides, inter alia, that the Commissioner may in writing determine the tariff headings, tariff subheadings or tariff items of any Schedule under which goods manufactured in the Republic shall be classified. Section 37(1) of the Act provides, inter alia, that excise duties are payable in respect of goods manufactured in a customs and excise warehouse, on entry for home consumption thereof at rates determined in terms of the Act.

colourants, and that they no longer qualify as a wine of any kind, but are ultimately spirituous and therefore liable to a tariff classification attracting higher duties.

[5] At the time of the determination, Distell assumed the position that the products in issue have a ‘basis of wine of fresh grapes’, are fermented, not distilled, and should resort under one or more of the following tariff headings, namely, 22.04, 22.05 or 22.06, all of which pertain to fermented beverages and consequently attract lower excise duties. Distell’s primary contention was that the products in question fell to be classified under tariff heading 22.04 in that they were ‘wine of fresh grapes, including fortified wines’. Alternatively, it contended that the products in issue are wine of fresh grapes (fortified wine), flavoured with plant and aromatic substances and accordingly, fell under tariff heading 22.05. It contended, in the further alternative, that the products are mixtures of fermented beverages and non-alcoholic beverages, as contemplated in tariff heading 22.06, which covers all fermented beverages other than those in tariff headings 22.03 to 22.05. Distell challenged the Commissioner’s determination that tariff heading 22.08 applies, as this heading does not, so it was contended, include aperitives ‘with a basis of wine of fresh grapes’.

[6] Subsequent to the Commissioner’s determination, set out in paragraph 3 above, Distell lodged an appeal in terms of section 47(9)(e) of the Act to the North Gauteng High Court3 on the basis set out in the preceding paragraph. That court (Pretorius J) found the products to be spirituous beverages and held that they thus fell under tariff heading 22.08. The present appeal is before us with the leave of the court below. We shall hereafter use ‘TH’ as an abbreviation for tariff heading.


3 Section 47(9)(e) provides that an appeal against any such determination shall lie to the division of the High Court of South Africa having jurisdiction to hear appeals in the area wherein the determination was made, or the goods in question were entered for home consumption.

[7] The Republic of South Africa is a party to the General Agreement on Tariffs and Trade and is a member of the World Customs Organisation, which

employs an internationally Harmonised System, referred to in the Act. Part 1 of Schedule 1 to the Act comprising the Section and Chapter Notes, the General Rules for the Interpretation of the Harmonised System and the tariff headings, is a direct transposition of the nomenclature of the Harmonised System.

[8] Section 47(8)(a) provides that:

The interpretation of–

  1. any tariff heading or tariff subheading in Part 1 of Schedule 1;

  2. (aa) any tariff item or fuel levy item or item specified in Part 2, 5 or 6 of the said Schedule, and

(bb) any item specified in Schedule 2, 3, 4, 5 or 6;

  1. the general rules for the interpretation of Schedule 1; and

  2. every section note and chapter note in Part 1 of Schedule 1,

shall be subject to the International Convention on the Harmonized Commodity Description and Coding System done in Brussels on 14 June 1983 and to the Explanatory Notes4 to the Harmonised System issued by the Customs Co-operation Council, Brussels (now known as the World Customs Organisation) from time to time: Provided that where the application of any part of such Notes or any addendum thereto or any explanation thereof is optional the application of such part, addendum or explanation shall be in the discretion of the Commissioner.’

[9] In the court below, Pretorius J started her reasoning leading to the conclusion referred to above by referring to the purpose of the correct tariff headings, namely, to determine the excise duty payable in terms of the Act. She considered TH 22.04, the relevant parts of which, together with their Explanatory Notes, read as follows:


4 Also referred to as the Brussels Notes.

Wine of fresh grapes, including fortified wines; grape must other than that of heading 20.09.

. . .

  1. Wine of fresh grapes

The wine classified in this heading is the final product of the alcoholic fermentation of the must of fresh grapes.

The heading includes:

  1. Ordinary wines (red, white or rosé).

  2. Wines fortified with alcohol.

  3. Sparkling wines. These wines are charged with carbon dioxide, either by conducting the final fermentation in a closed vessel (sparkling wines proper), or by adding the gas artificially after bottling (aerated wines).

  4. Dessert wines (sometimes called liqueur wines). These are rich in alcohol and are generally obtained from must with a high sugar content, only part of which is converted to alcohol by fermentation. In some cases they are fortified by the addition of alcohol, or of concentrated must with added alcohol. Dessert (or liqueur) wines include, inter alia, Canary, Cyprus, Lacryma Christi, Madeira, Malaga, Malmsey, Marsala, Port, Samos and Sherry.’

[10] In regard to this TH, Distell contended that, since it included wines fortified with alcohol, the beverages in question should continue to be regarded as fermented beverages, rightly resorting under this classification.

[11] As indicated, Distell relied in the alternative on TH 22.05, the relevant part of which, accompanied by the Explanatory Notes, reads as follows:

Vermouth and other wine of fresh grapes flavoured with plants or aromatic substances . . . This heading includes a variety of beverages (generally used as aperitives or tonics) made with wine of fresh grapes of heading 22.04, and flavoured with infusions of plant substances (leaves, roots, fruits, etc.) or aromatic substances.’

[12] The third alternative TH relied on by Distell was TH 22.06, the salient provisions and Explanatory Notes of which, are:

Other fermented beverages (for example, cider, perry, mead); mixtures of fermented beverages and mixtures of fermented beverages and non-alcoholic beverages, not elsewhere specified or included.

This heading covers all fermented beverages other than those in headings 22.03 to 22.05.’

[13] In contradistinction, the court below referred to the TH regarded by the Commissioner to be the appropriate one, namely 22.08, the applicable parts and Explanatory Notes of which, provide:

22.08 – Undenatured ethyl alcohol of an alcoholic strength by volume of less than 80 % vol; spirits, liqueurs and other spirituous beverages.

2208.20 – Spirits obtained by distilling grape wine or grape marc

2208.30 – Whiskies

2208.40 – Rum and other spirits obtained by distilling fermented sugar-cane products

2208.50 – Gin and Geneva

2208.60 – Vodka

2208.70 – Liqueurs and cordials

2208.90 – Other

The heading covers, whatever their alcoholic strength:

  1. Spirits produced by distilling wine, cider or other fermented beverages or fermented grain or other vegetable products, without adding flavouring; they retain, wholly or partly, the secondary constituents (esters, aldehydes, acids, higher alcohols, etc.) which give the spirits their peculiar individual flavours and aromas.

  1. Liqueurs and cordials, being spirituous beverages to which sugar, honey or other natural sweeteners and extracts or essences have been added (e.g., spirituous beverages produced by distilling, or by mixing, ethyl alcohol or distilled spirits, with one or more of the following : fruits, flowers or other parts of plants, extracts, essences, essential oils or juices, whether or not concentrated). These products also include liqueurs and cordials containing sugar crystals, fruit juice liqueurs, egg liqueurs, herb liqueurs, berry liqueurs, spice liqueurs, tea liqueurs, chocolate liqueurs, milk liqueurs and honey liqueurs.

  1. All other spirituous beverages not falling in any preceding heading of this Chapter . . . .‘

[14] The court below rightly held that it had to decide the meaning of the words in the various tariff headings, determine the nature and characteristics of the products in question, and thereafter select the most appropriate TH. In this regard Pretorius J referred to the following dictum in International Business Machines SA (Pty) Ltd v Commissioner for Customs and Excise 1985 (4) SA 852 (A) at 863G-H:

Classification as between headings is a three-stage process: first, interpretation – the ascertainment of the meaning of the words used in the headings (and relative section and chapter notes) which may be relevant to the classification of the goods

concerned; second, consideration of the nature and characteristics of those goods; and third, the selection of the heading which is most appropriate to such goods.’

[15] At this stage it is necessary to record, as did the court below, the proper approach to the consideration of tariff headings, Section Notes, Chapter Notes and Explanatory Notes. In Secretary for Customs and Excise v Thomas Barlow and Sons Limited 1970 (2) SA 660 (A) at 675D–676D, the following appears:

‘‘The duty which is payable is set out in Schedule 1 to the Act. This Schedule is a massive part of the statute in which all goods generally handled in international trade are systematically grouped in sections, chapters, and sub-chapters, which are given titles indicating as concisely as possible the broad class of goods each covers. Within each chapter and sub-chapter the specific type of goods within the particular class is


5 In terms of s 47(9)(e) an appeal against a determination by the Commissioner of a tariff heading is heard as a de novo application.

itemised by a description of the goods printed in bold type. That description is defined in the Schedule as a “heading”. Under the heading appear sub-headings of the species of the goods in respect of which the duty payable is expressed. The Schedule itself and each section and chapter are headed by “notes”, that is, rules for interpreting their provisions.

It is clear that the above grouping and even the wording of the notes and the headings in Schedule 1 are very largely taken from the Nomenclature compiled and issued by the Customs Co-operation Council of Brussels. That is why the Legislature in sec. 47(8)(a) has given statutory recognition to the Council’s Explanatory Notes to that Nomenclature. These Notes are issued from time to time by the Council obviously, as their name indicates, to explain the meaning and effect of the wording of the Nomenclature. By virtue of sec. 47(8)(a) they can be used for the same purpose in respect of the wording in Schedule 1. It is of importance, however, to determine at the outset the correct approach to adopt in interpreting the provisions of the Schedule and in applying the explanations in the Brussels Notes.

Note VIII to Schedule 1 sets out the “Rules for the Interpretation of this Schedule”. Para. 1 says:

The titles of sections, chapters and sub-chapters are provided for ease of reference only; for legal purposes, classification (as between headings) shall be determined according to the terms of the headings and any relative section or chapter notes and, provided such headings or notes do not otherwise indicate, according to paras. (2) to (5) below.”

That, I think, renders the relevant headings and section and chapter notes not only the first but the paramount consideration in determining which classification, as between headings, should apply in any particular case. Indeed, right at the beginning of the Brussels Notes, with reference to a similarly worded paragraph in the Nomenclature, that is made abundantly clear. It is there said:

In the second provision, the expression ‘provided such headings or Notes do not otherwise require’ (that is the corresponding wording of the Nomenclature) is necessary to make it quite clear that the terms of the headings and any relative section or chapter notes are paramount, i.e., they are the first consideration in determining classification.”

It can be gathered from all the aforegoing that the primary task in classifying particular goods is to ascertain the meaning of the relevant headings and section and chapter notes, but, in performing that task, one should also use the Brussels Notes for guidance especially in difficult and doubtful cases. But in using them one must bear in mind that they are merely intended to explain or perhaps supplement those headings and notes and not to override or contradict them. They are manifestly not designed for the latter purpose, for they are not worded with the linguistic precision usually characteristic of statutory precepts; on the contrary they consist mainly of discursive comment and illustrations. And, in any event, it is hardly likely that the Brussels Council intended that its Explanatory Notes should override or contradict its own Nomenclature. Consequently, I think that in using the Brussels Notes one must construe them so as to conform with and not to override or contradict the plain meaning of the headings and notes.’

[16] The court below went on to have regard to Rule 1 of the General Rules for the Interpretation of the Harmonised System, which states:

The titles of Sections, Chapters and sub-Chapters are provided for ease of reference only; for legal purposes, classification shall be determined according to the terms of the headings and relative Section or Chapter Notes . . . .’

[17] Pretorius J considered the Explanatory Notes to the Chapter Notes in relation to Chapter 22, under which the tariff headings in question reside. Those Explanatory Notes divide the products in Chapter 22 into four main groups, the relevant two of which are:

(B) Fermented alcoholic beverages (beer, wine, cider, etc.).

(C) Distilled alcoholic liquids and beverages (liqueurs, spirits, etc.) and

ethyl alcohol.’

It will be recalled that Distell contended that the products in question fall under category B, whereas the Commissioner determined that they fell under category C.

[18] The court below dealt with Distell’s contention that the products should be classified under TH 22.04, set out in paragraph 9 above, which refers to wine of fresh grapes, including fortified wines. Distell’s reliance on this TH was driven, inter alia, by the increased alcohol content of the products in question about which more will be said later. It will be recalled that the Explanatory Note to TH 22.04 states that the heading includes ‘wines fortified with alcohol’ and ‘dessert wines’.

[19] In this regard, the learned judge had regard to additional note 2 to Chapter 22:

The expressions “unfortified wines” . . . shall be taken to mean wine . . . with an alcoholic strength not exceeding 16 per cent of alcohol by volume and the expressions “fortified wine” . . . shall be taken to mean wine . . . with an alcoholic strength exceeding 16 per cent of alcohol by volume’,

He also referred to Explanatory Note (I)(4) to TH 22.04, the full wording of which is set out in paragraph 9 above. According to that note dessert wines are rich in alcohol and in some cases are fortified by the addition of alcohol.

[20] In deciding whether the contentions by Distell were justified, Pretorius J took into account the expert evidence of Dr Loubser (Loubser), a chemist and the Director: Quality Management and Research of Distell. In relation to dessert wines, Loubser testified to the effect that such wines are fermented and only alcohol or concentrated must, with additional alcohol are introduced to increase the overall alcohol content. Using the example of Madeira, which is a dessert wine, Loubser pointed out that no colourants, flavourants or sweeteners are added to create dessert wines.

[21] The court below considered the Commissioner’s submission that the products could no longer be classified as wine or fortified wine due to the fact that the wine had been stripped of the taste and flavour of wine and fortified by the addition of cane spirits to increase the alcohol content. The colourants, flavourants and sweeteners are then added and can thus be distinguished from dessert wines to which, as indicated above, no colourants, flavourants and sweeteners are added. Pretorius J sought assistance from a dictionary definition of wine which essentially describes a wine as an alcoholic liquor product from fermented grape juice. ‘Vinous’ is defined as being of the nature of/or resembling wine; made of or prepared with wine’.6


6 Taken from the New Shorter Oxford English Dictionary 6 ed (2007)

[22] Pretorius J then went on to cite a decision of the European Court of Justice, namely Siebrand BV v Staatssecretaris van Financiën [2009] EUECJ C-150/08. The court there was considering a case concerning a fermented alcohol-based beverage corresponding originally to TH 22.06, to which a certain proportion of distilled alcohol, water, sugar syrup, aromas, colouring and, in some cases, a cream base had been added, resulting in the loss of the taste, smell and/or appearance of a beverage produced from a particular fruit or natural product. The court held that this beverage did not fall under heading 22.06, but rather 22.08, as contended for in this case by the Commissioner. Although referring to the Siebrand case, Pretorius J considered this decision not to be binding on South Africa. For that, she relied on the decision of this court in the International Business Machines case, where the following appears (873J–874B):

Whatever may be the status of such a decision so far as customs administration and international organisations are concerned, it is not, until it is reflected in an Explanatory Note, authoritative in a South African Court. Before that, it is no more than an expression of opinion which involves the interpretation of the relative tariff headings and the Notes relating thereto.

Under our system, question of interpretation of the documents are matter of law, and belong exclusively to the Court.’

[23] Distell had submitted before the court below that the Explanatory Notes to 22.07, although not directly applicable, provided guidance in reaching a conclusion on the dispute in issue. The Explanatory Notes to 22.07 provides:

Ethyl alcohol is the alcohol which occurs in beer, wine, cider and other alcoholic beverages. It is obtained either by fermentation of certain kinds of sugar by means of yeast or other ferments and subsequent distillation, or synthetically.’

In juxtaposition are Explanatory Notes (A) and (B) to TH 22.08, which appear in paragraph 13 above. That deals with spirits produced by distillation and includes liqueurs and cordials.

[24] The court below had regard to a dictionary definition of ‘spirituous’,

being ‘of or pertaining to spirit or alcohol; containing (much) spirit or alcohol’.7 Pretorius J went on to consider Distell’s submission that TH 22.08 only has application to spirits produced by distillation and not by fermentation. According to Distell, the products in question are not liqueurs or cordials as set out in Explanatory Note (B) of TH 22.08, as they are not spirituous beverages. It is further provided that TH 22.08 does not include ‘(a) Vermouths, and other aperitives with a basis of wine of fresh grapes (heading 22.05)’. Thus, Distell contended TH 22.08 only applies to spirituous beverages and that, should the court find the products in question to be fermented beverages (as is their submission), TH 22.08 will not be applicable.

[25] Returning to the evidence by Loubser, Pretorius J considered his explanation that fermentation and distillation were two distinct processes and that distillation could lead to an alcohol content of 96 per cent per volume, while fermentation cannot be utilised to attain an alcohol content of more than 16 per cent. In both instances the alcohol contained in the products is ethyl alcohol. Furthermore, Loubser testified that a cane spirit is only added to the products in question to increase the alcohol content and the addition thereof does not deprive the wine of its character. Even when wine is fortified with spirits, the essential base character remains wine. Furthermore, by volume all the products in issue contain more wine than spirits and the wine component exceeds the spirit component (excepting Zorba). The absolute alcohol content of spirits in the products, excepting Brandy Ale, is higher than that of wine. Loubser, however, admitted that the wine is stripped of its taste and flavour, but did not explain the reason for so doing.

[26] The court below also took into account evidence on behalf of the Commissioner by Mr Michael Fridjhon (Fridjhon), an internationally recognised wine authority and wine judge and one of the country’s most

respected wine tasters and widely published wine writers. Fridjhon testified

about the organoleptic8 characteristic of the stripped wine. Fridjhon’s ___________________

7 Taken from the New Shorter English Oxford Dictionary 6 ed (2007).

8 Defined in the Concise Oxford English Dictionary 12 ed (2011) as ‘involving the use of, the

sense organs’.

conclusions were:

19.8.1 the residual aromas and tastes left in the wine after subjecting it to the stripping process are insignificant and would definitely not be discernible in the final product;

19.8.2 the perceptible difference between the stripped fortified wine and cane spirit diluted with water to approximately the same alcoholic strength is minimal . . . ’

[27] The court noted, on the basis of the evidence of Mr van Niekerk, the General Manager of Distell, that the wines used in the production of the products in question, were selected because they were low in flavour intensity, colour intensity, acid, phenolics and sulphur dioxide, and high in alcohol.

[28] The court below considered the Commissioner’s contention that the products in question should be classified under TH 22.08, the particulars of which appear in paragraph 13 above and more specifically that they resorted under subheading 2208.90, namely ‘other’. In this regard the court below had regard to the evidence on behalf of the Commissioner by Mr G Taylor (Taylor), who is a biochemist from the United Kingdom. According to him, the presence of spirits in the products in question was essential to obtain the required alcohol level and preserve it, as well as to add to the stability of added flavourants. Taylor, with reference to the evidence of Fridjhon, was of the view that it was not necessary to use the stripped wine as the same products could be produced by using neutral spirits as the alcohol base. The opposite was not true as the required alcohol strength could thus not be

obtained. The unique characteristics of wine were not required in the end product.

[29] Pretorius J stated that it was clear from the processes employed by Distell, which were demonstrated to and observed by Fridjhon, that the beverages in question were not only a mixture of a fermented beverage and

cane spirits, but that they were individually designed, each with a unique taste and characteristic. She held that the beverages in question consisted of several components, but that in each instance it was spirits that gave these products their essential character.

[30] The court below found that the Commissioner’s argument, that the alcohol component that gave the products in question their essential character was the spirits and not the wine, was well founded. Whilst concluding that all the products in issue were fermented alcohol-based beverages, Pretorius J nevertheless held that they can ‘by no stretch of the imagination’ be wines. The following appears in the judgement:

The addition of cane spirit, water, sweeteners, flavourants, colourants and cream in some instances, have caused new products to be created, which have lost all the aroma and taste of wine. Tariff Heading 22.04 can thus not be applicable.’

[31] Turning to the alternative classification, namely TH 22.05, the court below could not agree that it was an appropriate TH for the beverages in question. This conclusion was based on what she regarded as being common cause, namely that the products were not Vermouth and other wine of fresh grapes flavoured with plants or aromatic substances’. In this regard the court had regard to the Explanatory Note under this TH, which made it quite clear that the heading dealt with ‘a variety of beverages (generally used as aperitives or tonics) made with wine of fresh grapes of heading 22.04 and flavoured with infusions of plant substances (leaves, roots, fruits, etc.) or aromatic substances’. The addition of spirits, colourants, flavourants, sweetener and cream is not mentioned and thus, according to Pretorius J, this TH could never be the appropriate one.

[32] Referring to Distell Ltd v The Commissioner, SARS [2011] 1 All SA 225 (SCA), Pretorius J held that the beverages are produced in a multiple stage process – two beverages are not mixed to get the relevant product. The colourant, flavourant and sweetener mixture cannot be described as ‘lemonade like’ or ‘cooldrink like’ (as Distell contended), does not constitute a non-alcoholic beverage and thus could not fall under one of the ‘mixtures’ referred to in TH 22.06 which is set out in paragraph 12 above.

[33] Finally, the learned judge concluded that the wine in the products in issue does not contribute to the organoleptic characteristics of the final products as it is neutral and cannot give it its essential character. Accordingly, the court found that all of the products in issue are spirituous and resort under TH 22.08 and, more particularly, under TH 2208.90.20.

[34] Thus it is the correctness of the reasoning and the conclusions set out above that are at issue in this appeal.

[35] Before us, reliance on TH 22.04 was abandoned by Distell. Its case in the present appeal is that two of the beverages in question, namely Zorba and Brandyale, fell under TH 22.05 and the remaining 13 under TH 22.06. The reason for this distinction, so they contended, was because, in the case of the former two products, water was not added, and they could thus not be considered to be mixtures as contemplated in TH 22.06.

[36] It is now necessary to follow the approach set out in the International Business Machines case, described in paragraph 14 above. First, we have to interpret the tariff headings concerned. Starting with TH 22.05, it is clear that this TH refers to wine which is the fermented product derived from fresh grapes. The Explanatory Note states that the beverages under this heading include a wide variety of beverages (generally used as aperitives or tonics) made with wine of fresh grapes of TH 22.04 and flavoured with infusions of plant substances or other aromatic substances. It is clear that what is dealt with in this paragraph is a product derived through the fermentation process to which fresh grapes are subjected, with plants or aromatic substances being added to the fermented liquid.

[37] It was Distell’s case that the addition of spirits does no more than ‘fortify’ the stripped wine used in the making of the beverages. TH 22.04, so it was contended, provides for the fortification of wines of fresh grapes by way of the addition of alcohol in whatever form. According to Distell this fortification process does not in any way change the essential vinous character of the base of stripped wine. Following that logic, Distell submitted that the two products in question, therefore, resided more logically and appropriately under TH 22.05.

[38] On behalf of the Commissioner it was contended that the base of stripped wine was no longer wine and that this liquid could, even if alcohol be added to it, not qualify as fortified ‘wine’, as none of the base liquid’s essential vinous qualities were retained. Moreover, they submitted that the ingredients added at the end of the process can hardly be described as being ‘flavoured with aromatic substances’.

[39] We now turn to consider TH 22.06. This TH covers all fermented beverages other than those provided for in TH 22.03, TH 22.04 and TH 22.05. TH 22.06 refers to ‘[o]ther fermented beverages (for example, cider, perry, mead); mixtures of fermented beverages and mixtures of fermented beverages and non-alcoholic beverages, not elsewhere specified or excluded’.

It was common cause that the beverages in question do not fall within the genus under which cider, perry and mead reside. We were required to consider whether the beverages were mixtures of the kind contemplated in this TH. The mixtures that are contemplated are clearly of a combination of fermented beverages or of fermented beverages with non-alcoholic beverages added, which do not properly reside under any other TH.

[40] In respect of the remaining 13 products, Distell contended in relation to TH 22.06 that these products were mixtures of fermented beverages and non-alcoholic beverages. The non-alcoholic beverage on which Distell relied, is the mixture of water and flavourants, sweeteners and colourants. Distell argued that TH 22.06 does not require that a mixture of a fermented beverage (eg fortified wine) and a non-alcoholic beverage should retain the character of a particular type of fermented beverage, for instance wine. Furthermore, they argued, that whatever the processes the wine was subjected to, in order to reduce it to an almost wholly neutral alcoholic liquid, it still retains its essential character, namely, of wine. Lastly, Distell contended that, in any event, the mixture resulting in the products is not spirituous in character, in that the volume of the stripped wine is greater than that of the cane spirits, except for one of the products, and they submitted that the mixture in itself does not have the essential characteristics of spirits.

[41] In making the argument referred to in the preceding paragraph, Distell submitted that one could not argue, as the Commissioner does, that what we were dealing with in relation to the products in question was a mixture or combination of a once fermented beverage with a distilled beverage. In order to counteract the Commissioner’s contention in this regard, Distell was driven to submitting that the addition of the cane spirits was merely a fortification of the existing stripped wine. In this sense, so it was submitted, one was dealing with a fortified wine which on its own was undoubtedly a fermented beverage to which the non-alcoholic components, which flavoured, coloured and sweetened the beverage, together with the water were added.

[42] It is now necessary to have regard to the evidence about the nature of the beverages in question. The parts of Fridjhon’s evidence, referred to in paragraph 26 above, were dealt with by Loubser, as stated hereafter. Loubser’s response was not to contest that the flavour and aroma of the stripped wine is negligible. Loubser adopted the position that a fermented product such as wine can only change its ‘essential character’ when distilled and not when subjected to the processes in question. However, in Loubser’s founding affidavit the following is stated:

Wine is selected for its sensory and analytical characteristics.’

This is in line with Fridjhon’s primary assertions. In Fridjhon’s answering affidavit he refers to the Oxford Companion of Wine, in which flavour is said to be ‘arguably a wine’s most important distinguishing mark’. Fridjhon went on to state that vinosity is the defining element of wine.

[43] The evidence of Taylor, referred to in paragraph 28 above, is important. Loubser’s evidence concerning a fortified wine such as Madeira, in support of Distell’s case, is unhelpful. It is true that Madeira, a fermented product, has brandy, which is a distilled product, added to it to increase its alcohol content. Fridjhon’s responding affidavit makes it clear that like all recognised fortified wines, the addition of spirits does not cause Madeira to lose its essential vinosity. On the contrary, its vinosity is bolstered by the addition of spirits.

[44] It was common cause that the stripped wine’s maximum alcohol content was between 12.5 per cent and 16 per cent, the latter of which is recognised as a general maximum for an unfortified wine. The addition of the cane spirits increased the alcohol content to between 18 per cent and 23 per cent.

[45] Another important part of the evidence on behalf of Distell is that the production sequence in relation to the beverages ultimately produced was unimportant. More particularly the stripped wine could have been added at the end of the production process.

[46] It is clear from the evidence that the wine was subjected to the stripping process to neutralise its taste and aroma. Final fermented products, even in the case of fortified wines, do not lose their essential vinous characteristics. Much as distillation changes the essential characteristic of a fermented product, so too do the processes which result in the stripped wine. The following question posed by Taylor illustrates the point:

If, as is argued, these are wine based products and the wine is an integral component, why then is the base wine neutralised? If the wine character is that important, then surely it should be retained and the fortification be utilised to enhance that character and help carry it into the final product? The fact that the wine character is removed prior to fortification strongly suggests not only that the wine character is not required, but that it is actually undesirable.’

[47] In our view, Distell’s reliance on the overall volume of the stripped wine in relation to the cane spirits is misplaced. Clearly, one could have a greater volume of water overwhelmed by a lesser volume of an intense different liquid. It is a question of which essential ingredient is dominant. In this regard General Rules of Interpretation 3(b) provides that in the case of mixtures, the goods are to be classified as if they consisted of the material or component which gives them their essential character, in so far as this criterion is applicable.

[48] It is now necessary to revisit TH 22.05, set out in paragraph 11 above. As stated earlier, the essential characteristic of a beverage resorting under this TH is that of a ‘wine of fresh grapes’. For the reasons set out in the preceding paragraph it cannot be, in our view, said that the stripped wine forming the basis of the two beverages in question qualifies as wine under this TH, for the reasons provided by Fridjhon and Taylor and due to the common cause facts mentioned above. As Fridjhon, supported by Taylor and Dr Croser, the wine maker who also testified on behalf of the Commissioner, pointed out:

What such processes would have removed would have been precisely what fermentation contributed in the first place: the essential vinosity of the product. The restoration of the alcohol to the fluid left after the flavour and alcohol had been removed would not thereby produce wine . . . ’

[49] Distell’s contention that, even though the stripped wine has lost much of its flavour and aroma, it is nevertheless a fermented product and a wine is, in our view, for the reasons stated above, fallacious. Consequently, the two products in question do not fit under TH 22.05.

[50] Turning to the remaining 13 beverages, we now reconsider TH 22.06. In our view, Distell’s reliance on this TH is also unjustified. An essential requirement of this TH, for the purposes of Distell’s argument, was that the fermented beverage used in the production of the products was fortified wine (‘wine’ in the sense of TH 22.04). As we have already demonstrated, the ‘stripped wine’ cannot be regarded as wine for the purposes of TH 22.04, and therefore cannot be made ‘fortified wine’ in the sense used in TH 22.06. Furthermore, a fortified wine does not itself lose any of its vinous qualities and it appears that, if anything, the vinosity is thereby enhanced. That is not the case with the beverages in question. The fact that the sequence of production is irrelevant demonstrates further that the submission by Distell is unsustainable.

[51] Following on the conclusions reached in the preceding paragraphs it follows that the next enquiry is whether the beverages in question rightly resort under TH 22.08, which is set out on paragraph 13 above. It is clear, when one has regard to the TH, that the beverages do not resort under tariff sub-heading 2208.20, in that they are not spirits obtained from distilling grape wine or grape marc. It is common cause that they do not fall under any of the other tariff sub-headings between 2208.30 and 2208.70. It is equally clear that they cannot be classified under tariff notes (A) or (B). As set out above, the cane spirits was added to the stripped wine to boost alcohol content significantly. According to Taylor, he had tested all 15 beverages organoleptically and concluded that they all have a distinct spirituous character. Considering our line of reasoning set out above, in relation to the beverages in question, and in particular paragraph 47, the compelling conclusion is that the ultimate distinctive nature of the beverages is spirituous, that they rightly resort under TH 22.08, and are covered by tariff note (C).

[52] Distell’s reliance on the decision of this court in Distell Ltd and Another v Commissioner for SARS [2011] All SA 225 (SCA) is misplaced. In that case it was common cause that TH 22.06 applied. The dispute was whether the beverages fell under the first or second part of the item. It was submitted on behalf of Distell that there was no difference to the facts of this case in that the ‘wine coolers’ in issue in that case constituted wine, to which flavourants and water had been added. It was submitted that the vinous nature of the ‘wine coolers’ were not challenged in that case and that in the present case, neither should the vinous character of the beverages in question. It was submitted on behalf of the Commissioner that, contrary to this case, there had been no attempt in the first Distell case to mask the flavour of the wine by a stripping process. We agree that the facts of that case are poles apart from those in the present appeal.

[53] We were referred by the Commissioner to another decision by the European Court of Justice (ECJ), namely Paderborner Brauerei Haus Cramer KG v Hauptzollamt Bielefeld [2011] EUECJ C-196/10. In that case, the ECJ was called upon by the Finanzgericht Düsseldorf to make a preliminary ruling on whether ‘a liquid described as a “malt beer base”, such as that in issue in the main proceedings, with an alcoholic strength by volume of 14% and obtained from brewed beer which has been clarified and then subjected to ultra-filtration, by which the concentration of ingredients such as bitter substances and proteins has been reduced, must be classified under tariff heading 2208 of the CN’.

[54] The ECJ found that the ‘malt beer base’ was not a beverage for the following reasons. Although suitable for human consumption in the sense that it was drinkable, it was not an end product primarily intended for consumption, but rather an intermediate product for use in the production of another product; the malt beer base was not sold to consumers as an end product; it was not obtained purely and simply by fermentation, but was after fermentation subjected to ultra-filtration which caused it to lose its ‘objective properties and characteristics particular to beer’. The Explanatory Note to TH 22.08 expressly states that the heading also covers ethyl alcohol, whether intended for human consumption or for industrial purposes and, although this Explanatory Note excludes from that heading alcoholic beverages obtained from fermentation, the malt beer base, not being a beverage, was not affected by the exclusion. Finally, the fact that the malt beer base was not completely devoid of any aroma did not exclude it from being classified under TH 22.08. The malt beer base, after being treated, was ethyl alcohol and as a consequence must be classified under TH 22.08.

[55] Counsel for the Commissioner submitted that, like the malt beer base, the stripped wine is not produced purely and simply by fermentation; is devoid of the vinous character of wine of fresh grapes; is not sold to customers as an end product; is an ‘intermediate product’ specifically ‘prepared’ to be used, and used, in the production of the products in issue, and that it satisfies the requirements of the Explanatory Note proviso to TH 22.08. Thus, following the analysis and interpretation of the ECJ in this case, the stripped wine is not ‘wine’ as contemplated by TH 22.04 and would be classifiable under TH 22.08.

[56] We were warned on behalf of Distell to be cautious about the dangers of relying on decisions by the ECJ. According to counsel, the ECJ had simply made a ‘preliminary ruling’ concerning the interpretation of the combined nomenclature of the common customs tariff. The main proceedings were before the Düsseldorff Court. This ‘preliminary opinion’ is a non-binding opinion, the admissibility and status of which should not be over-emphasised. Moreover, counsel contended the Parderborner case does not support the Commissioner’s contentions. The treatment of wine does not change the essential character of wine, and the Commissioner did not lay any factual foundation why the process used in the Parderborner case (ie to treat the malt beer base by ultra-filtration) is comparable to the processes used by Distell in respect of the wine it used in the manufacturing of the beverages in issue.

[57] None of these submissions is convincing. Clearly the decisions of the ECJ are not binding on South African courts. They may have persuasive force, but it is up to the South African court to decide the relevance of the foreign decision in question. It was also not necessary for the Commissioner to demonstrate that the processes followed in the Parderborner case were identical to those followed by Distell in relation to the beverages in question.

[58] Whilst the conclusions in Parderborner and Siebrand accord with our own, we have arrived at our decision by applying the Harmonised System as catered for by the Act and following the line of logic and reasoning set out in the preceding paragraphs.

[59] In the result, the appeal is dismissed with costs, including the costs of two counsel.








FOR APPELLANT: A.P. Joubert SC (with him C Louw)

Instructed by

Shepstone & Wylie Attorneys


Webbers Attorneys


FOR RESPONDENT: C.E. Puckrin SC (with him J.A. Meyer SC and I A Enslin)

Instructed by

The State Attorney


The State Attorney