[This is a guest post by Anshul Dalmia.]
In the recent case of Arvind Kejriwal’s plea for recusal, Justice Sharma uses the ‘real danger’ test to deny judicial disqualification (¶ 186). The High Court states that the petitioner had to show the existence of a real, direct, and substantial connection or interest which may affect the impartiality of the Judge. This connection or interest must not be assumed, imagined, or inferred from remote circumstances but should rest on tangible material and a clear nexus with the issues before the Court (¶ 192).
The self-evolving jurisprudence on recusal law by the Supreme Court has led to the creation of several tests throughout the years namely, the real likelihood test, the real danger test and the reasonable suspicion or apprehension test. The Courts have conveniently applied these tests in various cases to answer questions involving the issue of judicial disqualification. This blog builds on the earlier two posts (here and here) to showcase the skewed legal tests and standards governing recusal within the Indian constitutional jurisprudence. The ‘real danger’ test imposes a high burden on litigants who find it difficult to tangibly prove an inherently mental element such as bias. I seek to highlight the additional issues underlying the ‘real danger’ test and push for the adoption of a calibrated ‘reasonable suspicion’ test.
Tracing the Shift From ‘Real likelihood’ to ‘Real Danger’
The real likelihood test dominated the jurisprudence surrounding the recusal framework for a very long time. In most cases, the Courts were unable to strike a balance between their constitutional duty to hear the case and the constitutional right of the litigant for a fair trial, giving supremacy to the former over the latter. This test was whether the judge is likely to be biased in his perspective. Thus, this was considered from a reasonable man’s perspective as to who will be presumed to have complete knowledge of all the circumstances which showcase the plausibility of bias. However, the circumstances of the case should be cogent enough to allow the bias to be readily inferred. This implies the presence of non-ambiguous facts, which in itself is a rare occurrence, especially in cases which are adjudicated by multiple judges in several States, in different forums and various court-like structures.
The major shift from the real likelihood test came in 2016, in the five-judge bench case of Supreme Court Advocate-on-Record Association v. Union of India, where Justice Chelameswar introduced and evolved two different tests, namely the ‘real danger’ and the ‘reasonable suspicion’ test, based on the landmark case of R v. Gough (‘Gough’).
The Gough case adopted the ‘real danger’ test as the dominant applicable touchstone on which recusal orders need to be passed. This test entailed disqualification solely on substantive and tangible evidence which conclusively highlights the presence of judicial bias and prejudice. The judges in the Court of Appeal believed that the ‘real danger’ test would be the remedy to all issues of discrimination and help achieve ‘pure’ justice. According to the Court, this test was more conclusive in its results as it would avoid the permeation of doubts. The Court did not conclusively lay down the boundaries of this test but attempted to define its contours by differentiating it with the real likelihood test.
The introduction of a ‘reasonable observer’ standard posed as an additional burden for the courts since they had to be the one behind the personification and subsequently adjudicate the pertinent likelihood of bias. Moreover, this test would exclude the creation of ‘relevant circumstances’, which are determined through a holistic consideration of all the available evidence that hint towards the bias. Mere facts of the case would not allow the observer to ascertain the probative value of the evidence placed for adjudication. The subjective interpretation of injustice would allow a sense of bias to be perpetrated, whereas the actual circumstances on the ground could be very different. Thus, it was imperative for the Court to bridge the gap between these tests to prevent the impression of discrimination being based on suspicion or apprehension. However, even after all these criticisms, the Court of Appeal went one step further. It highlighted that the ‘reasonable suspicion’ test could be used in cases when there is an inquiry apart from allegations of pecuniary bias. This implied the preservation of the ‘real danger’ test.
The Supreme Court of India applied a congruent approach to conclude that in an inquiry, not including allegations against pecuniary interest, the judges could use either the ‘real danger’ test or the ‘reasonable suspicion’ test, but emphasised on the former (¶ 26). In the following part, I attempt to explore the issues spanning the dominance of the ‘real danger’ test in the jurisprudence surrounding the recusal framework and suggest the endorsement and subsequent enforcement of the ‘reasonable suspicion’ test.
Moving Beyond: From ‘Real Danger’ to ‘Reasonable Suspicion’
I argue that the adoption of the ‘real danger’ standards and the convenient exclusion of the ‘likelihood test’ raises several concerns. The primary issue is the choice provided to the judges to apply either of the tests in cases of judicial disqualification. It can be deduced that the main reason behind the inculcation of the ‘real danger’ test was to add a blanket of protection onto the judiciary and also make it onerous to raise questions challenging the bias of a judge.
Though the specific contours of the real danger test have not been conclusively carved out, the need of turning to actual evidence to prove substantial bias has been delved into. The shift for proving recusal of judges from the balance of probabilities to a burden of proof which is beyond doubt showcases the likelihood of the ‘real danger’ test being prioritised over an effective alternative, i.e., the ‘reasonable suspicion’ standard. The Supreme Court Advocate-on-Record Association case, being a five-judge bench, is the latest binding precedent and the entrenched focus on the ‘real danger’ test highlights dangerous times ahead.
A recusal petition is primarily looked at as questioning the impartiality of a judge and challenging his conscience to deliver a fair decision. This in itself is an arduous task as only circumstantial evidence may be gathered which in turn may hint towards presence of a bias. No evidence can be collected of the discrimination present in the mind of a person. Moreover, the types of influences on a person’s mind may be so innate that it is practically difficult to delineate them as a bias. These internal features prevent the presentation of actual or real evidence pointing at tangible circumstances. Additionally, it is difficult for anybody in the legal community to identify and discern the point of permeation of bias in the process of decision-making. Accordingly, a very high burden is placed on the litigants, forcing them to compromise their perception of justice. Thus, there is an emerging need for the ‘reasonable suspicion or apprehension’ test to be accepted in its entirety.
The judges in Gough emphasised on the ‘real danger’ test while being ignorant of the point of view of the common people. The judges, while laying down the tenets of the test, highlighted the need to satisfy them, the possibility of bias and discrimination. The Court categorically mentions that it would be ‘unsatisfactory’ that in a recusal framework, the ‘public point of view’ or the petitioner’s perspective are considered. With the increasing need to reinstate the public confidence in the judiciary for litigants to respect this dispute resolution process, it is pertinent that a shift is made from the real danger test and the accompanying standards. The former Chief Justice of India, MN Venkatachaliah, while laying down the tests for recusal, categorically held that the proper approach in answering questions of bias must involve looking at the parties involves and their perception of discrimination. Hence, a shift is required from a paradigm where the courts are introspecting the presence of a possible bias to a framework where the judges consider the perceptions of both the litigant and the common people.
It is pertinent to note that there is an inconsistency between the real likelihood test and the reasonable suspicion test, even though they seem to espouse common notions. Admittedly, both the tests use the reasonable man standard in ascertaining the extent of bias. However, the major difference is in the standards of ascertaining bias. The former is based on the likelihood of bias, whereas the latter test is founded on the apprehension of bias. It can be noted that the qualifying terms highlight the standards of disqualification. For instance, ‘likelihood’ implies a probability greater than fifty percent, whereas suspicion highlights a ‘low exacting test’.
Criticisms of the ‘Reasonable Suspicion’ Test
I posit that the ‘reasonable suspicion’ test would eventually lead to the creation of a ‘fair minded, informed and reasonable observer’. The House of Lords in Gough had criticised this theory based on the inability of the observer to segregate the relevant facts of circumstance from the plethora of evidence made available. This process was the duty of the judge, who was adept and qualified to undertake the segregation. Thus, the test applied an unnecessary burden on the observer. However, this position of the Court is streaked with several issues.
Firstly, it is onerous to disassociate the position and creation of the ‘reasonable man’ in a situation where the judges themselves adjudicate the dispute. It has been observed that courts move towards the creation of this fiction for two reasons. First, to pay attention to the objectivity of the law and, second, promote transparency by looking at the individual qualities of the parties. Additionally, the creation of an egalitarian society was sought for by reinstating equality and fairness. In order to move a step towards transparent decision- making processes, an objective criterion must be present, which would act as the touchstone of recusal orders.
Secondly, the judges might not always be independent in categorising circumstances which need subsequent filtration for a reasonable observer. Every factual interpretation by the judge is affected by several forces of influence, either consciously or subconsciously. These deep rooted biases implicitly allow the judges to be selective in taking note of relevant facts or conform to their pre-existing notions of the law. Additionally, cultural prejudices, social biases, and psychological structures prevent the permeation and subsequent assimilation of all the evidence and the facts. Thus, this reasoning seems counter-intuitive, as there will always be an innate possibility of a bias in the fact collection and filtration processes. Hence, this is a flawed approach due to the inherent contradiction that stems from discarding the reasonable man standard in favour of a method that itself suffers from inherent bias.
Finally, the House of Lords itself corrected their view and stance on the exclusion of the ‘reasonable man’ standard and included the above standard in a subsequent decision. The Court tweaked the test by splitting it into two prongs – the first step would be ascertaining the real tangible circumstances. Only if they were proved, then the concept of the ‘fair-minded and informed observer’ had to be introduced to answer questions of judicial disqualification. However, most of the recusal requests fail to bypass the first stage due to the strict level of scrutiny. With the problems elucidated above relating to the adoption of the ‘real danger’ test coupled with the affinity of the Courts, I suggest the sole incorporation of the ‘reasonable suspicion’ test within the Indian jurisprudence. While it is clear, that some issues highlighted in Part I would be present here as well, the ‘reasonable apprehension’ test seems to have several effective safeguards.
The major safeguard present with respect to this test is the qualifying need for it to be ‘reasonable’. This qualification would, thus, alleviate the concerns of the court regarding the increase in frivolous objections. The concept of even including the modified ‘reasonable man’ standard would allow for the integration of impartiality in the judicial framework. The alleged suspicion must also relate directly to the issue at hand in such a manner that it could prevent the decision-maker from reaching a fair decision. An ascertainment of both the proximate nexus and cogency with the issue at hand must always be undertaken.
I argue that this test is markedly different from the real danger and likelihood test as it bases its foundation on the ‘reasonableness’ of the suspicions raised. This test allows us to harmoniously reconcile the court’s problem with this framework as well. The concerns relating to the possibility of bench hunting might be genuine in very few cases. However, the entire test must not be vindicated solely based on a ‘once in the blue moon’ probable occurrence. The judiciary must impose a higher level of scrutiny and be careful in dealing with cases which allege bench selection, rather than outrightly dismissing every recusal petition in lieu of these concerns. The imposition of safeguards coupled with holistic legislation might be the panacea to the malady of forum shopping. While proposing this test, I conclude by reemphasise the positive duty of courts to preserve and maintain the confidence of the public in the judicial mechanisms.
It is imperative that the constitutional right of a fair trial is protected. Even the probability of bias must lead to the elimination of a judge. In an adjudication procedure, where no discrimination is expected, the probability of a bias must be dealt with utmost care. Even the slightest probability of prejudice would lead to a complete vindication of public confidence in the mechanism of the judiciary. The court’s current attitude towards this issue has been one of outright dismissal and subsequent ignorance. Hence, even the mere likelihood of discrimination must be assessed fairly.