(Lord McGhie, Mr A Macdonald, Mr J A Smith)
(Application RN SLC 136/03 – Order of 13 October 2011)
EXPENSES – CROFTS – CROFTERS RIGHT TO BUY – RIGHTS OF ACCESS – DISPUTED TERMS AND CONDITIONS – MOTION FOR REHEARING – PROPER APPROACH TO EXPENSES – UNREASONABLE CONDUCT – INEFFICIENT CONDUCT OF LITIGATION – OFFERS TO SETTLE – CROFTERS (SCOTLAND) ACT 1993 SEC 12–15
This case had a long procedural history in course of which matters of expenses were reserved. The crofter applied to buy his croft but faced difficulties in respect of the access which ran over land which had been conveyed to a third party with certain rights of access reserved to the croft landlord. The third party changed the route. The applicant unsuccessfully contended that he had not agreed the change. Shortly before the proof, the third party and landlord entered a Deed of Conditions restricting the access rights. The applicant would not accept the more restricted conditions. Several days were devoted to hearing and submissions on these issues. The decision of the Divisional Court was partly overturned on appeal. The landlords presented a motion for rehearing which was refused. The Court dealt with expenses on the basis of written submissions.
The Note appended to the Court’s order is as follows:
 This case has a long history in course of which matters of expenses were reserved. Parties are now agreed that we should deal with this issue before further procedure in relation to the special case requested by the third respondents. They were also agreed that we should deal with the matter on the basis of written submissions. We refer to our Note of 5 September 2011 for some detail of the basis upon which we proposed to proceed. Put shortly, the applicant seeks an award of expenses against the second respondent and third respondents, jointly and severally, or severally. The second respondent and third respondents each move for expenses against the applicant on an agent and client, client paying basis. The fourth respondent also moves for expenses against the applicant. Extensive written submissions and counter-submissions were lodged.
 A summary of the substantive facts of the case can be found in our Note of 4 December 2009. The following is a bare outline of the procedures followed. The application was lodged in December 2003. The Divisional Court heard evidence on 4 and 5 June 2006. The first day was largely spent in negotiations in relation to access to the croft. There was an understanding that this had been agreed and there was reference to a Joint Minute being lodged. The hearing on 6 June was largely confined to questions arising in relation to the landlords’ reservation of access over the croft. No Joint Minute ever materialised and a procedural hearing of January 2007 revealed no more than that there was a dispute which would require proof. The hearing of evidence took place on 25 and 26 April with submissions being heard on 29 June 2007. The Court, by Order of 12 September 2007, then invited written submissions on one specific issue before finally issuing a decision of 21 December 2007. We heard an appeal by the third respondents against part of that Order on 14 and 15 October 2009 and issued a decision on 4 December 2009. Following that decision, there was some delay before the applicant eventually lodged an amendment in suitable terms seeking an Order against the second respondent. Following the amendment, the second respondent accepted that he should grant a servitude of access over his land in favour of the applicant and on 17 August 2010 there was a hearing to discuss the terms and conditions of access over the areas of land belonging to the second and third respondents. On 1 February 2011 we held a hearing on the third respondents’ motion to allow a rehearing in relation to one of the conditions we had ordered in relation to access. The second respondent did not require to attend that hearing.
 In this short summary we have not attempted to deal with the substantive outcomes of the various procedures because these are at the heart of the disputed motions and require more detailed consideration. It should also be said that although we intend the expenses of preparation of submissions on expenses to be covered by our order, the expenses of and relating to the third respondents’ request for a special case are not the subject of the present dispute
 Before dealing with the detail of the present case, it is appropriate to make some general observations about the proper approach to expenses. The Court does have a wide discretion and it is well established that this is to be exercised on a broad basis. We do not require to carry out a detailed examination of all aspects of the case. But it is equally clear that our discretion must be exercised in accordance with established principles. It is not disputed that the main principle to be applied in relation to expenses is that expenses should follow success. That is a convenient shorthand for matters more fully expressed by Lord President (Robertson) in Shepherd v Elliot (1896) 23 R 695 at 696: “The principle upon which the Court proceeds in awarding expenses is that the cost of litigation should fall on him who has caused it. The general rule for applying this principle is that costs follow the event, the ratio being that the rights of parties are to be taken to have been all along such as the ultimate decree declares them to be, and that whosoever has resisted the vindication of those rights whether by action or defence, is prima facie to blame. In some cases, however, the application of the general rule would not carry out the principle, and the Court has always, on cause shown, considered whether the conduct of the successful party, either during the litigation, or in the matters giving rise to the litigation, has not either caused or contributed to bring about the law suit”.
 Although the latter reference to the conduct of the successful party might suggest a wide test, we are satisfied that it is unreasonable conduct in relation to the litigation which is to be looked at having a bearing on expenses. When considering expenses, the Court is not concerned with issues of conduct which are properly the focus of dispute on the merits. That may be a narrow issue in some circumstances.
 In considering the “success”, the straightforward approach will usually be to look at the substantive effects of the Orders of the Court and compare these with an applicant’s craves or conclusions. However, it is clear that this is not a rigid rule. The Court is entitled to look at other circumstances. In some courts there are formal rules about tenders. But even where an offer in settlement does not take the form of a tender, we have no doubt that in finding an answer to the question of who has caused the expense of litigation to be incurred, the court is entitled to have regard to the terms of offers made by either party to settle the dispute.
 The proper construction of offers is an important issue in the present case. Put shortly, the respondents rely on an analysis of the applicant’s pleadings to show that it was not until a very late stage that he made clear what Order he wanted from the Court. His original craves had been unsuccessful. The applicant, in essence, says that the respondents were well aware of what he wanted and that that was broadly what he got at the end of the day.
 We shall return to the question of reasonable conduct. Both sides make assertions of unreasonable conduct by the other. However, for the purposes of the present case, we proceed on the basis of a distinction between unreasonable conduct of a litigation and unreasonable belief in the soundness of an argument or position. If it can be said that someone has advanced an argument purely for tactical reasons and knowing it has no substance, that can be described as unreasonable conduct of the litigation. Both sides complain of aspects of the other’s conduct before the hearing of any evidence. Both assert that they were justified in the steps they took. It is recognised that to reach a conclusion on these matters might have required a substantive proof in itself. Having agreed that the matter be dealt with on written material, it is clear that we could only make a finding based on unreasonableness if this was clearly demonstrated by unambiguous and unchallenged written material. As will be seen, we have not been persuaded that there is a proper basis for such a finding in this case.
 It is well established that a broad approach to expenses is appropriate. A party who has achieved overall success may have deployed some unsuccessful arguments in course of the litigation. These will normally fall to be ignored when considering liability for expenses but, occasionally, when it can be seen that unsuccessful arguments have caused significant, identifiable, extra expense, it may be unreasonable for the other side to be expected to pay for them. We prefer to characterise that as inefficient conduct of the litigation rather than as unreasonable conduct. A typical example is where a party has been unsuccessful at debate. It is quite usual to find that party liable in expenses even where it cannot be said that he has acted unreasonably in taking a particular argument. Such findings can be justified because the failure at debate demonstrates that the point has not contributed to the efficient progress of the litigation.
 We do not think that, in the context of expenses, it can be said to be unreasonable conduct to fail to reach agreement. The proper course for a respondent faced with an apparently unreasonable opponent is to spell out his best offer. If this is not accepted, the terms of that offer can be compared with the effective orders of the Court at the end of the day. Something similar may be done by an applicant who is prepared to accept less than his formal crave. However, different issues arise where he indicates that he is prepared to accept something other than his crave. A party may well be entitled to resist an unfounded claim even if that resistance does not avoid a similar end result. This might be an uncertain area where the claim and the alternative would involve the same party. We need express no view. Essentially, the difficulty for the applicant in the present case is that the second and third respondents were different parties against whom different claims were being made. The primary approach to the question of success in a litigation is to see what the parties were contending for and compare that with the ultimate award. In considering questions of expenses, the start point is that parties are normally entitled to rely on the pleadings. If a party seeks to rely on other material it will be for them to satisfy the Court that it provides a secure basis for assessment.
 We shall return to the pleadings. But we think it appropriate, in the present case, to start by taking a look at the broad overall picture. We recognise that there is a sense in which the applicant may appear to have been a victim of a deliberate plan to deprive him of his normal rights. It is necessary for us to address that issue. As will be seen we have no doubt that there was such a plan but use of the word “victim” may be misleading in the context of assessment of liability for expenses of the litigation.
 The applicant was a crofter. He had a right, which was not disputed before us, to buy his croft. As a crofter he had access for pedestrian and vehicular access over a clearly defined route over lands which had formerly belonged to his croft landlord but which had come to be in the ownership of the second respondent, third respondents and fourth respondent respectively. This access was not subject to any explicit conditions or restrictions although it may be noted that it was implicitly restricted to his use as a crofter. These rights were commensurate with the rights of access which the landlord of the croft initially held. Had the purchase proceeded in the normal way, the applicant would have acquired good title to the access from his landlord. The applicant would have acquired from the landlord broadly the same rights as he enjoyed as tenant. The Court would have imposed specific conditions as to maintenance and, perhaps, some limit on the nature of use, to ensure that the parties over whose land access was taken, did not suffer any significant increase in the burden. These issues would not have required elaborate pleading. They would fall within a familiar compass. It is seldom that anything advanced at a hearing in relation to such conditions would take anyone by surprise. The matter would have been disposed of without any great procedural difficulty or expense.
 The protracted dispute over the applicant’s right to access on purchase became prolonged and complicated by two factors. One was that, the applicant having agreed to variation of the access route within the second respondent’s land and having allowed physical work to be carried out in reliance on that agreement, came to deny it when difficulties arose in respect of his rights relating to the use of that altered route. The second was a dispute over the nature of rights of access to be given in relation to the second respondent’s ground. Shortly before the first proof hearing, the third respondents and second respondent freely entered an agreement which, in the language of the third respondents’ submission on expenses, “compromised” the landlords’ rights of access. Instead of access without significant restriction the third respondent accepted a range of restrictions. Accordingly, he was no longer able to give a good title to include the rights he had held at the time when the crofter first sought to acquire them. It is important to note that there was no attempt by the crofter to challenge the Deed of Servitude by action of reduction. His position in the present case was that the actings of the respondents in entering such a Deed amounted to unreasonable conduct and that note of this should be taken in assessing liability for expenses.
 There seems no doubt that the acceptance by the third respondents of the restricted conditions of access was intended to impose a restraint on the use of the access by the applicant. It is not clear whether it was originally thought that the restrictions would affect him as tenant but they were agreed at a time when it was clear that his intention was to buy the croft. The main aim must have been to impose restrictions on the applicant as purchaser.
 For present purposes, it is to be accepted that the second respondent and third respondents had good reasons for trying to restrict the applicant’s use of the access and that these reasons related to prior history of use of the access. It is impossible to say that a desire to impose restrictions on the access was in itself unreasonable. We heard various submissions properly advanced in relation to conditions or restrictions of access at the hearing in August 2010. The respondents were entitled to try to take any other legal steps they could to achieve that end. The fact that any particular steps taken might turn out to be ineffective would not necessarily mean that it had been unreasonable for them to take them. It may be that within the line of authority sometimes referred to loosely as “the Rodgers v Fawdry cases” there would have been found some basis for challenge of the Deed. But that would not have been a matter for this Court. There was no such challenge and, for our purposes, the Deed existed as simply part of the material to be considered on the merits. The second and third respondents were entitled to advance such contentions as they thought proper, based on the existence of the Deed of Servitude.
 In other words, we do not accept that any question arises as to any sanction in expenses based on the reasonableness or otherwise of the second and third respondents’ attempts to restrict access by creating a Deed of Servitude with new terms and conditions. The questions raised as to the significance of such a Deed can be seen to be a part of the disputed issues to be dealt with by proper procedures. The fact that the existence of the Deed made it more difficult for the applicant to see the way forward, does not mean that the attempts made by the second or third respondents to protect their own interests were necessarily unreasonable. They were entitled to take what steps they could to protect their perceived interests.
 There is, of course, no doubt that the existence of that Deed and the restrictions imposed by it, introduced much of the complexity in the case. Initially there was an agreement that the Servitude would be withdrawn but, once it proved impossible to reach agreement on other aspects, the offer to withdraw was not repeated. Much of the time taken in submissions – including the stage of the written submissions requested by the Court’s Order of 12 September 2007 – was taken exploring the implications or consequences of the Servitude.
 In the circumstances it is necessary to take a careful view of the question of whose actings gave rise to the litigation. Plainly litigation would have been needed to deal with the detailed conditions of access quite apart from the Deed. The hearing in August 2010 disclosed a variety of such issues relating to the route as a whole although focused by the arguments presented by the third respondents. In other words, it is not possible simply to look at the pleadings, or the case as a whole, when trying to determine the question of overall success. It is necessary to consider why matters followed the course they did. This requires us to approach the matter by looking at the various stages of the litigation.
 The applicant accepts that the second and third respondents may be said to have enjoyed a measure of success at specific steps of procedure. But he contends that the Court has made an Order against them broadly in terms of those applied for at the start or “agreed” by the applicant before evidence was led. We shall return to look at the detail of the craves but it is plain that the Order is not broadly in terms of what was applied for at the start and the submission that it was in terms agreed by the applicant is, in itself, indicative of a lack of clarity. An applicant is a person applying to the Court for an Order. His agreements, as such, are irrelevant. The issue is what he seeks. No doubt the submission is intended to mean that he had shown, by his willingness to agree certain things, what he really was asking the Court to do. Had he spelled that out in his pleadings, much trouble might have been saved.
 The main contention is that much of the procedure could have been avoided had the respondents acted reasonably in their conduct of the action. The submission goes on to refer to the “considerable significance” of the Deed of Servitude. Reference is also made to the obvious close co-operation between the second and third respondents. But the applicant does not spell out the significance of either of these factors in the context of expenses. If his position was that the respondents were not, as a matter of law, entitled to enter the Deed of Servitude, he could have taken steps to have it reduced once he realised that it was not to be withdrawn voluntarily. He did not do so and the Court had to deal with matters on the basis that the Deed of Servitude was effective to do what it purported to do: that is, give restricted rights of access to the owner of the Knapp. The applicant makes reference to the tract of correspondence between the hearing in June 2006 and the hearing in April 2007 and draws from it the conclusion that it shows that the second respondent was intent on requiring the applicant to accept the conditions of the Deed of Servitude. The second respondent appears to dispute this and submits that there was “a clear and binding undertaking” (presumably binding on the second respondent) to withdraw the Deed. The present reliance on this is hardly consistent with the many submissions made about that Deed. Had the second respondent made clear to the Divisional Court his view that the Deed of Servitude should be treated as set aside, there would have been no need for much of the argument heard by us and by that Court. In any event, we accept that the correspondence shows that the applicant’s position was that he would accept the varied route if the second respondent would grant a servitude over that route in terms and conditions to be set by the Court as opposed to the terms and conditions in the Deed of Servitude.
 But the formal application to the Court did not make it clear that the applicant was seeking any Order against the second respondent beyond the declarator relating to the route. Had he had a crave against the second respondent, the parties would have had to address it and the Court would have had to deal with it. Instead, the applicant felt “compelled to assert his right of security of tenure in the existing pertinent of access”. This led to the unfortunate attempt to dispute the agreement to the change of route. It may be added that it is not clear that there was, at any stage, a dispute as to the security of his tenure in the pertinent of access over the new route. When we raised that question at the start of the appeal hearing, we understand it to be agreed on all hands that the applicant’s rights of unrestricted access as tenant were not in dispute. The question was the nature of rights to be granted to him as a purchaser.
 The applicant also contends that the third respondents’ position in reliance on the Deed of Servitude as the beginning and end of the argument was not upheld. However, we think that submission is a misunderstanding of our decision on the appeal. We accepted Mr Thomson’s submissions as largely well founded: . We did not suggest that his client could be compelled to act in a manner inconsistent with a title certificate incorporating the terms of the Deed of Servitude. Our substantive decision related to the position of the second respondent.
 The applicant characterises the respondents’ reliance on the Deed of Servitude as intended to frustrate the applicant in his attempt to purchase his croft. Whatever the truth of that, it is not self evident and we have nothing to support it. The dispute over access would not prevent purchase. The issue was the nature and extent of the rights to be granted as access to the croft. The applicant’s pleadings did not properly focus that issue. His pleas did not address the matter until after the appeal.
 The applicant placed some reliance on the close relationship between the respondents and, indeed, the Divisional Court also commented on this. But it seems to us that the fact of this close relationship and that it may have led to counsel for the third respondents advancing arguments in support of the second respondent’s position, has led to a failure by the applicant to keep in mind that they were different parties with interests in different areas of land. Put simply, it is not success in an action against A to gain an award against B. It is essential to look at the respondents as separate parties.
 In procedural terms, we think that the main problem created by the Servitude was that it obscured the real issues. The applicant required to direct his attentions to overcoming the very real difficulties it created for him. It now seems clear that his attempts to get round the problem were in some respects reasonable but ill-founded and in others disastrously wrong. The creation of formal rights over the new route may have been impeded simply by inability to agree some detail of terms and conditions. The second respondent now makes reference to some dispute about issues of title or ownership but these are not reflected in the terms of the decision of the Divisional Court and we, ourselves, heard no submissions bearing directly on that matter. It is not clear precisely where the problem arose although it does seem clear that it led to the second respondent refusing to grant any servitude over the land. It is very unfortunate that, when agreement could not be reached, the applicant took the view that his only way forward was to deny any agreement to change the route. He did not amend to seek to have the Court determine his rights over the route. He sought to enforce his rights by reliance on the rights he had over the original route. The Court did not accept his evidence in relation to the agreement to vary the route. It is impossible to ignore the fact that a very substantial portion of the two days of evidence and day of submission in 2007 was taken up to deal with the question of change of route. The applicant was unsuccessful and, apart from other considerations, as this involved sharp issues of credibility, we think it clear that he must be found liable in the expenses of the three days of hearing involved.
 For the respondents it was contended, in effect, that the very fact that the Court did not accept the applicant’s evidence must, in the circumstances, mean that he was to be treated as a deliberate wilful perjurer and that this justified a finding that expenses be taxed on an agent and client basis. We do see some force in the argument. However, it is one which might be advanced in many cases and we are not aware of any practice that a person whose evidence is not accepted should be found liable in a penal rate of taxation. It must be recognised that parties’ recollections are imperfect. Memory can be distorted by strong feelings. The applicant was plainly frustrated by the impasse over his access. He had good reason to be. He may well have persuaded himself that the discussions had not been as clear cut as the Court ultimately accepted them to have been. In any event, in the whole circumstances of the present case, we are satisfied that taxation should be on a normal basis.
 Before the Divisional Court, the applicant sought a remedy against the third respondents and that was what the Court purported to give. This led to the appeal and was an issue in respect of which the applicant was clearly unsuccessful. This would not necessarily have led to a finding of liability in expenses against him. The appeal was simply a step in the process of enforcing his right to buy his croft. The existence of the Servitude gave rise to difficult and complicated questions of law as is noted in the third respondents’ submissions: at paragraph 13.3. The issues caused by the third respondents’ compromised title were not only difficult and complex but it seemed that the situation which the respondents had created was one without direct precedent. We would not necessarily have been persuaded that the failure of the applicant’s arguments on this point should be divorced from his ultimate success and treated as a separate issue had we been dealing with a case against one respondent. However, we think it impossible to ignore the fact that the substantive effect of the appeal was that the third respondent was successful and freed from any question or liability to grant or procure the grant of a Servitude over the second respondent’s subjects. The appeal clarified the issue of who could grant the servitude and as the third respondents were shown to be the wrong parties we think it necessary to follow the normal course of finding the applicant as losing party, liable in the expenses.
 Although the Land Court has always sought to avoid undue reliance on the formalities of pleading, there are some disputes where the issues are complex and a need for precision is clear. It is usually assumed that the crave will be the measure of what the Court is being asked to do. In this case, the difficulty of identifying the nature of the right of access to be given to the applicant and the precise terms and conditions to be attached to any grant on purchase was, or should have been, apparent by the end of 2006. The applicant made no proper attempt to set out in his pleading the nature of what findings the Court was being asked to make to meet the disputed matters at that time. The proof over three days in 2007 was a battle over something which was not truly in dispute: whether the access should run on the old route or the new one. Had the issue been focussed properly on the question of who could be compelled to grant rights of access over the new route, how such rights should be constituted and the conditions to be attached, much of the time and effort involved in the hearings in June 2007, and indeed, in the appeal might have been avoided. As it was, the argument advanced successfully by Mr Henderson on behalf of the applicant at the appeal, was not advanced before the Divisional Court. When we look closely at the pleadings the gulf between them and the issues which the parties actually required the Court to determine is quite apparent.
 In the first crave the applicant sought declarator that he held, as a pertinent of his croft, a right of access over a route shown green on the plan appended to the application. That was the original route. The applicant was plainly unsuccessful on that issue. The issue under the second crave related in effect to areas to the east of the peninsula and that was conceded by the applicant, as we have seen. The third crave sought an order authorising the applicant to acquire land “delineated in red”. This was the peninsula area. It did not include any access track. The craves made no reference to such track. The fourth crave related to the house site and has not been in issue to any extent.
 The applicant refers to negotiations and in effect contends that the written craves should be taken to have been varied by the terms of his letters. Before looking at the detail of this, it is necessary to say a little more about the third respondents’ submission that it was a fundamental principle that a court could not determine expenses on the basis of parties’ extra-judicial negotiations. That is plainly an important issue in the context of the present case. We invited submission in support of it and the third respondent provided a full and helpful note. We are satisfied that it would be inappropriate for a court to attempt to hold a proof on the fine detail of discussions. However, we are equally satisfied that if a clear and unambiguous offer can be identified and if it can be said that, at the end of the day, the result did not better that offer, this would be a factor properly to be taken into account in relation to expenses. In short, we do not accept that there is any principle that a court cannot determine expenses on the basis of extra judicial negotiations. The question is one of circumstances and requires assessment to determine how clearly any particular party’s position has been brought to the attention of the others and how it is to be related to the litigation itself. As we have said an offer to take something in substitution for the crave may present some difficulties and may not be relevant when different parties are involved.
 It appears that the parties were fighting over different issues at different stages of the case. We can properly look at the correspondence to see whether any clear positions were there set out. It may be said that we have not found it possible from the terms of the correspondence lodged, to identify the real reasons for failure to reach agreement. The issues which took on the main importance in subsequent stages were the attempt to have the third respondents grant rights over the second respondent’s land and, later, the question of whether the applicant should be limited to access to one dwelling. The latter was an issue which might have been expected in any similar type of case. It may, indeed, have been the real point in dispute but it was never mentioned in the pleadings and was not focused as a dispute in the correspondence.
 It may be said, for completeness, that the second respondent makes some reference to a dispute as to land where there was a complication of title involving the Scottish Ministers. It is suggested that this was an issue which prevented agreement. It does not appear that this was referred to in any detail before the Divisional Court and it does not appear from the correspondence lodged that this was a matter preventing settlement. We do not think that anything of significance now turns on this.
 One of the third respondents’ submissions makes reference to the applicant’s “intransigence” in seeking an Order against the third respondents. But there is nothing to suggest that the applicant’s position ever changed from that set out in his solicitor’s letter of 16 June 2006. He offered to give up the old route. He wanted the second respondent to grant a servitude for access for vehicles and pedestrians over that part of the re-routed road passing through his property, expressed in terms similar to the way in which the right was reserved in the second respondent’s own title at that time (that is before effect was given to the Deed of Servitude). He asked that the fourth respondent grant a Servitude in similar terms. That is what he eventually obtained.
 It is of interest to note that the second respondent did not expressly reject the idea of a grant of a servitude. Indeed a peculiarity of the whole matter is that in the formal pleadings the second respondent expressly offered to grant a servitude right of access on the terms of the original reservation to the landlord. Nothing was made of that offer in written submissions nor proceedings before us. It may have been assumed to have been departed from by the subsequent Deed of Servitude. However, in his letter of 27 August 2006 the second respondent’s solicitors wrote: “I have been instructed to reach some agreement with the landowner to the south of Mr Doble’s property, Mr MacColl, as to the terms of the Servitude to be granted in favour of the crofter”. This plainly showed a recognition that a sensible way forward was for the owners of the burdened lands to grant the relevant rights. In his letter of 7 March 2007, Mr Henderson referred expressly to the onus on Mr Doble to grant a replacement servitude. He sent a draft joint minute which included provision for the Deed of Servitude to be withdrawn. The second respondent’s response was to raise various technical legal objections. It was said that he was in favour of granting to the applicant a right of access “as a pertinent of his croft” but not as a servitude right. He contended that agreement to drop the Deed was inappropriate as the Court had no power to order this. We think that there can be no real doubt that the second respondent was well aware, before the hearing in April 2007, that the applicant was seeking a title which included a grant of a new servitude of access by the second respondent. But, equally, the second respondent was aware that he did not face any risk of an award by the Court in such terms as there was nothing in the craves which could have led to such an order.
 When the question of access came before the Court at the April 2007 hearing, there had been no attempt by the respondents to withdraw the Deed of Servitude. Their simple position was that any rights given to the applicant on his purchase of the croft had to recognise that the third respondents’ rights as owners were restricted by the terms of the Deed of Servitude. The third respondents could not be compelled to grant a wider right. That created a difficulty for the applicant. He thought the third respondents had been part of the agreement to withdraw the Deed. However, the position of the respondents was that no binding agreement had ever been reached. They were, accordingly, free to take a stand on the legal position as it stood in light of the Deed of Servitude. The applicant was unable to contend that matters had reached the stage where there was a binding agreement.
 It is clear that the applicant pressed on with a contention that the third respondent should grant or procure rights of access over the second respondents land. Much time was spent on that argument. It was unsuccessful.
 On the question of conditions of access, the position is less clear. It is not suggested that either of the respondents ever indicated that they were willing to grant access for more than one dwelling. There are some indications that the applicant would have agreed to be limited to two but nothing clearly to that effect and at the hearing about conditions he did not set that out as his position. For the record, it may be noted that the indications come, first, from a letter of 2 June 2006 where his solicitor said: “Mr Gourlay does not accept as reasonable a restriction of access to one house, especially considering that there are two dwellings on the croft at present”. But, perhaps, a more telling indication can be found in the letter of 11 October 2006 from the fourth respondent’s solicitor (which appears to have crossed Mr Henderson’s letter of the same date). He referred to a telephone conversation with Mr Henderson. He said that the fourth respondent would not want to see an unrestricted right of access “but will have no difficulty, I suspect, in agreeing possibly one additional or replacement house to what there is at present”. That suggests that the writer knew that the applicant was looking for something on these lines. However, this issue, as between the applicant and the second and third respondents, was never clearly focused as a dispute between one dwelling or two dwellings.
 The applicant seems to suggest that he was successful because his position was that the Court should simply determine terms and conditions of purchase. However, if terms could be agreed, they should have been. If they could not be agreed, the obvious inference must be that one side was claiming or asserting something which was not acceptable to the other. If the Court was left to determine the matter it would need to know what a party’s respective contentions were. So this proposition must be construed as meaning that on some issues there was a disagreement which the applicant wished to have resolved by litigation. As he does not say what his position was in respect of all such issues it is not possible from the material now available to us to tell whether the Court’s ultimate findings were more or less advantageous to any one party than the offer made by the applicant. We can, of course, have regard to the competing submissions at the hearing.
 Some of the points made by the second respondent do seem indicative of a positive desire to place obstruction in the way of agreement. For example, in response to the letter of 7 March 2007, when the applicant proposed that if suitable rights were established over the new route, the old route could be “resumed”. It was disputed that this was appropriate. The second respondent was not a landlord and could not resume the right. However, it must have been obvious that the substantive offer was that the rights over the old route should be extinguished. Dispute over the technical language simply got in the way of identification of the real issue. However, this simply brings us back to the problem that, faced with the difficulties created by the Deed of Servitude, the applicant did not have a clear view of the remedies he required. That Deed certainly created a situation of novelty and difficulty but it is for an applicant to identify the remedy he seeks and take efficient steps to allow the Court to determine whether he should be found entitled to that remedy. It does not follow that every inefficient step taken in the course of a litigation will necessarily have a bearing on expenses. In the present case, while we find little excuse for the attempt to insist in the old route, we cannot easily blame the applicant for the difficulty he faced in identifying the proper solution.
 In answer to the applicant’s contention that he has been successful if the Court’s findings are taken as a whole, the second respondent contended that the applicant had not been successful in terms of his original craves. That is plainly correct although it does not meet the substantive argument. We do not think that it is a distortion of events to say that the outcome was on the lines of the proposed Joint Minute sent by the applicant with his letters of 7 March 2007. But it is important to note that there were other issues and it is too simplistic to suggest that as the second respond ought to have known what the applicant was really wanting he should have disregarded the pleadings and conceded that. The second respondent contended that any question of terms and conditions was a matter between the applicant and the third respondents. There was said to remain a conflict between an offer of access “as presently exercised” and “unrestricted access”. This would have been an oddly imprecise basis over which to conduct a litigation and this imprecision is another aspect of the problem of a lack of explicit pleadings.
 It ultimately appeared that, quite possibly, the main real dispute related to use for one or more houses. Although it is possible that the applicant might have accepted access for two houses if offered, there is no attempt to suggest that he had ever made any offer to accept such a restriction. At the hearing on conditions in August 2010 Mr Thomson made it very clear that the third respondents were seeking a restriction to use for one house. Mr Henderson did not suggest that his client’s goal was simply to ensure that he would be able to build and have access to an additional house for his own use. His argument was couched in terms of “unrestricted access” although he did make reference to the possibility of restriction “for crofting purposes”. In relation to the applicant’s essential requirement to have a full right of access to cross the second respondent’s land free of the restrictions in the Deed of Servitude he can be said to have been successful but he was not successful in having this free of all conditions. In relation to the important question of the number of houses to be served by the access, no party can be said to have been successful. (We note for completeness that the proposed shared maintenance was conceded from an early stage.)
 In summary, it can be said that this was a very unusual case. The main complication was the agreement of the respondents to enter the Deed of Servitude. That complication was, in itself, made more complicated by the initial agreement, discussed in Court on 6 June 2007, to withdraw the Deed, followed by their decision not to do so but to rely on it as a critical factor in relation to the rights to be given to the applicant on his purchase. (If, as Mr Duncan now contends, the respondents were aware that they had entered a binding agreement to withdraw the Deed of Servitude, their conduct of the hearing was inexplicable. Much time and effort would have been wasted by the failure to make this clear. But we simply do not accept that this is an accurate account of the second respondent’s position) We are aware of the strength of the applicant’s feeling that it would be wrong and unjust to find the respondents entitled to expenses in this case when the Deed of Servitude was plainly the cause of so much trouble and when he was ultimately successful in obtaining a right of access unrestricted by it. However, perhaps because the applicant regarded the granting of the Deed as unreasonable conduct, he did not face up properly to the fact that the respondents were entitled to advance arguments based on it.
 We recognise that although much of the expense of this case would be taken in preparation for hearings there would also be a good deal of expense in relation to negotiation and routine processing of the case. It is apparent from our consideration of the whole lengthy submissions of parties in relation to the expenses and consideration of the whole terms of all the various notes by the court and the correspondence now founded on that there are many factors to balance. It may be said that our initial impression was that the tenant had indeed achieved broadly what he would have been expected to achieve from the outset and we have some sympathy for his position overall. We had some difficulty in identifying precisely what it was that led to the failure to reach agreement. However, we have no doubt that we must have regard to the various steps of procedure actually taken by the applicant. In a litigation, respondents are entitled to respond to the specific claims of an applicant and we have no doubt that the primary approach is that described by the Lord President in Shepherd v Elliot. We cannot simply substitute an impression of fairness.
 At the end of the day it is clear that the applicant did not achieve any success against the third respondents in respect of the hearings in 2007 and the appeal. He did have some success in relation to the question of conditions to be applied to the rights of access to and over the croft. That success was mixed.
 As between the applicant and the third respondents, we are satisfied that, in light of the findings of the Divisional Court, the applicant must bear the expense of the hearing which started on 25 April 2007 including the written stage which followed. We accept that the third respondents were wholly successful at the appeal and that there is no good reason why the applicant should not be liable in respect of their expenses at that stage. Similarly it is clear that the third respondents were unsuccessful in relation to their motion for rehearing and it is not disputed that the applicant should have the expenses of that procedure. These awards will necessarily carry the expenses of preparation and other expenses of and incidental to attendance at the hearings and the expenses of the written submissions.
 Apart from these expenses identified by reference to the hearings and the rehearing, we have taken the view that it is appropriate to find no expenses due to or by, as between the applicant and third respondents. We reach that view on a broad overview of the application. The applicant was entitled to bring an application to determine his right to purchase and the conditions to be attached to it. He had to face the difficulties created by the Deed of Servitude and apart from his denial of the change of route we cannot say that any of the other efforts he took to deal with the situation were of a character which required special treatment in expenses. At the end of the day there was mixed success on the detail of the conditions to be attached. In respect of the hearing in June 2006, the applicant’s success in relation to conditions of access over the croft which was the main matter disputed at that hearing was matched by his concession in relation to the second crave. There was an apparent agreement in relation to the first crave leading to the Court being told that a Joint Minute would be lodged. It is inappropriate now to attempt to find any party responsible for the expenses of the issues which were so resolved at the time. In respect of the hearing of August 2010 relating to the disputed conditions, there was mixed success in relation to the minor conditions and in relation to the main dispute over the number of houses to be served by the access, neither party can be said to have had success.
 The position as between the applicant and the second respondent must be viewed separately. It is plain that most of the running at the substantive hearings was made by Mr Thomson who saw the interest of the third respondents as extending to cover the arguments advanced by the second respondent. We accept that the third respondents were entitled to be present at the hearing although their interest in the choice of route was technical rather than one of any real substance. Had the issues not been addressed by Mr Thomson they would have had to be addressed by Mr Duncan. Although we have had some concern over the second respondent’s contention that he knew all along that he would be obliged to withdraw the Deed of Servitude, we think that a confusion. We accept that essentially his position in relation to the 2007 hearings and submissions was the same as the third respondents. He must be found entitled to the expenses.
 In relation to other matters it can be said that the second respondent was successful on a technical view of the pleadings. However, there can be little doubt that the substance of the dispute was the question of whether or not the applicant was entitled to access over his land free from the restrictions created by the Deed of Submission. Mr Thomson’s submissions were emphatically to that effect and Mr Duncan simply adopted Mr Thomson’s contentions on most matters and occasionally attempted to amplify them by short submissions of his own. At no time before the applicant’s amendment in 2010 was any attempt made by Mr Duncan to spell out in Court that his client was prepared to grant a right of access on new terms. At the end of the day the applicant was successful. He had not adopted the most efficient way of achieving that success but we cannot characterise his general conduct of the litigation as unreasonable when the respondents, themselves, have pointed to the novelty and difficulty of the issues he faced.
 It was not strictly necessary for the second respondent to be present at the appeal hearing. The appeal was against an order directed at the third respondent. In relation to the second respondent a substantive issue addressed at appeal was the question of whether an order could be pronounced against the second respondent under the provisions of section 16. The applicant was successful in that argument. But, it had not been foreshadowed in his written submissions prior to appeal and the second respondent had not had time to prepare to meet it. At the hearing Mr Duncan was offered time to consider matters further. Fortunately, he said he was able to deal with the matter without requiring a continuation. In the circumstances we do not think it appropriate to find the applicant entitled to expenses against the second respondent in respect of the appeal. In any event his substantive expense at that hearing would be occasioned by the third respondent’s appeal. As between applicant and second respondent we think it appropriate to find no expenses due to or by either party.
 For the reasons discussed in relation to the third respondents we also conclude that apart from the expenses of and relating to the substantive hearing in 2007, it is appropriate in relation to the second respondent to find no expenses due to or by. On the substantive issue of the terms to be imposed in relation to the applicant’s right of access over his land, there was divided success in the proper sense of that term.
 The submission on behalf of the fourth respondent can be shortly stated. He sought an award of expenses on the basis that he lodged a Minute of Sist in terms similar to the Final Order of the Court. However, it is not clear that he required to do so or that he required to incur the expense of attendance or involvement in any subsequent stages of the case. Formally, the dispute related to the route through Mr Doble’s land. That, of course, had implications for the continuation of the track through the fourth respondent’s ground but it is not apparent that he had any interest to resist the applicant’s first crave. He gave evidence restating his position that he would grant a Deed of Servitude for the new route. As there was no apparent issue or dispute between him and the applicant it cannot properly be said that the conduct of the litigation by the applicant caused him to incur expense. Of course, he may well have incurred expense as a result of the various negotiations about access. But that is another matter. In the whole circumstances we think it appropriate to find no expenses due to or by the fourth respondent.
 The third respondents seek sanction for employment of counsel. However, we do not think it appropriate in all the circumstances of the present case. There is no doubt that it gave rise to difficult and novel issues but these were almost entirely due to the Deed of Servitude entered into by the respondents in the middle of the proceedings. Most of the time at the hearing before the Divisional Court was taken by evidence on the question of the choice of route through the second respondent’s land. Although much of the running was conducted by Mr Thomson, we are not satisfied that the third respondents had any proper interest in that and, in any event, the main issue was the factual one of whether or not the crofter had given consent to the change of route. At the appeal stage, the third respondent’s position was one of some general importance: was a croft landlord to be compelled to procure rights over the second parties land when he had willingly given up or compromised his own rights as owner? However, Mr Thomson’s position was that it was a straightforward issue and the appeal was not seriously disputed by the applicant.
 This is a matter for the discretion of the Court. Our rules refer simply to the “difficulty and general importance” of the case. Many crofting cases involve a client’s home or livelihood and that is implicitly a factor to which we can usually give some weight. In the present case the third respondents have no such direct interest. The circumstances of this case were unusual rather than of general importance. It is obvious that the other parties did not consider it necessary to instruct counsel - although Mr Henderson might have had more justification for doing so. In all the circumstances we think it in accordance with justice to refuse the motion for certification.
 We are satisfied that it is appropriate to make payment of expenses a condition precedent to the purchase. We have no doubt that it would be wrong for the third respondents to lose what is the heritable security of the croft by being compelled to convey it before the debt is paid. The position of the second respondent is not quite the same but we accept it as reasonable in principle that he should not be compelled to grant a valuable right in favour of a debtor until the debt is paid. It may be added that we recognise that there may be a need as part of the formal conveyancing procedures to have this condition expressed in a more technical way. In case there is any difficulty over this we reserve leave to any party to seek a more precise order.
For the applicant: Mr S. Henderson, Solicitor, Portree
For the second respondent: Mr T Duncan, Solicitor, Linlithgow
For the third respondents: Mr D Thomson, Advocate; Brodies, Solicitors, Edinburgh