Loughland, Amelia --- "Female Judges, Interrupted: a Study of Interruption Behaviour During Oral Argument in the High Court of Australia" [2020] MelbULawRw 10; (2020) 43(2) Melbourne University Law Review 822

	Home \| Databases \| WorldLII \| Search \| Feedback Melbourne University Law Review

An interruption of speech, such as that experienced by Kiefel J, violates the ‘conversational contract’[2] by disrupting the implicit rule of turn-taking in any dialogue.[3] Such interjections are not simply impolite, but allow the interrupter to assert dominance over their conversational partner.[4] Behavioural research has found that men tend to interrupt women at a far higher rate than vice versa,[5] suggesting that conversational interruptions both reflect and maintain unequal gender power relations in society.[6] Yet it would be reasonable to expect that such interpersonal expressions of dominance would be suppressed in settings where women hold institutional positions of power over men. This expectation is supported by American studies that have found, in an experimental setting, that a speaker’s position of authority may have a greater effect on speech patterns than gender.[7]

However, this article will reveal how women even at the pinnacle of a high-status career continue to experience gendered behaviour. Descriptive statistical analysis of interruptions as recorded in transcripts of High Court oral argument shows that female High Court judges were interrupted far more frequently than their male colleagues. These findings echo those of Professor Tonja Jacobi and Dylan Schweers in the US, who showed that gender was the salient explanatory factor for interruptions between judges across three terms of the US Supreme Court, even after taking into account ideological and hierarchical divisions on the Court.[8] As well as finding empirical evidence for a systemic gender effect in interruption behaviour between 2015–17, I also found that interruptions of female judges counterintuitively increased in 2017 when the Court was first led by a female Chief Justice, the Hon Susan Kiefel AC. The fact that everyday conversational patterns are replicated in the highest courts is surprising when considering how far removed formal oral argument is from the setting and norms of everyday conversation, and illustrates how gendered norms transcend even the highest spheres of judicial authority.

In this article, I consider possible explanations for the gendered patterns of interruption of judges in oral argument, including their respective volubility, seniority and manner of speaking. A closer look at the specific kinds of interruptions reveals how gendered conversational norms are reproduced by male advocates against female judges,[9] and how their deferential behaviour towards male judges further may indicate an institutional bias in favour of masculine judicial authority. I then consider the normative implications of these results for the participation of female judges on the High Court and question whether greater representation of women on the Bench is enough to ensure their substantive equality. Ultimately, I argue that advocates must be made aware of their implicit biases and that the Chief Justice should take a more active role in regulating oral argument. However, given that interruptions increased under Kiefel CJ, a male Chief Justice is less likely to be seen as ‘biased’ towards other women and thus may be more effective in this regulatory capacity.

II The Empirical Study

To determine whether there was a gendered effect in interruptions of High Court Justices and the impact of a female Chief Justice, I analysed transcripts from oral argument in all Full Bench hearings from June 2015 (when Gordon J joined the Court) until the end of 2016 and compared these findings to those from 2017, when Kiefel CJ had replaced French CJ as Chief Justice. These transcripts are publicly available on the High Court’s website,[10] and average about 60–70 printed pages in length.

Given that Edelman J was appointed as puisne justice at the start of 2017, there was a consistent gender distribution (three female and four male) and a roughly equal number of hearings (22 in 2015–16 and 23 in 2017) across the two periods. Full Bench hearings were selected in order to maintain a consistent gender balance in each case that was analysed in the dataset. For the purpose of counting interruptions (coded by ‘- - -’ in the transcripts),[11] I was agnostic as to the interrupter’s purpose, and assumed that each interruption had a disruptive impact on the judge speaking at the time.[12]

A Findings

In 2015–16, the three female judges collectively received 52% of the total number of interruptions, whereas the four male judges received 48%.[13] While I expected the interruptions of female judges to decrease under a female Chief Justice in 2017, they in fact increased: the three female judges received 69% of the total interruptions, whereas their four male counterparts received only 31%.

As evident from Figures 1 and 2, the Chief Justice in both periods received the highest number of interruptions. In 2015–16, French CJ was a clear outlier among the male judges, demonstrated by the fact that Nettle J was interrupted the second most often but represented only 9% of interruptions overall compared to French CJ’s 26%. Yet French CJ’s interruption rate is very close to that of Kiefel CJ in 2017, who received 30% of the total interruptions. This suggests that, because the Chief Justice intervenes to regulate oral argument in their ‘chairperson’ capacity,[14] they have a higher chance of being interrupted by advocates.

Accepting that the role of Chief Justice requires atypical judicial behaviour and requires their removal from the dataset, the gendered interruption pattern becomes stark. After excluding French CJ from the 2015–16 dataset, the three female judges still received 71% of the remaining interruptions (receiving 115 of the total remaining 163 interruptions),[15] despite the fact that they are now being compared with an equal number of male judges. Similarly, even after removing Kiefel CJ from the dataset in 2017, interruptions of the two female judges still represent 56% of the total (receiving 77 of the remaining 138 interruptions), meaning they were interrupted at more than twice the rate of their male colleagues.

III Possible Explanations

A Volubility

The simplest explanation for this disparity is that women create a greater opportunity for interruption by speaking more than their male counterparts during oral argument. However, by counting each judge’s total number of speech episodes,[16] I found that they spoke roughly in proportion to the representation of their gender on the Court.

Speech episodes	Total (Percentage)
By female judge (3)	1573 (37.2%)
Kiefel J	806 (19.1%)
Bell J	449 (10.6%)
Gordon J	318 (7.5%)
By male judge (4)	2653 (62.8%)
French CJ	1374 (32.5%)
Gageler J	474 (11.2%)
Keane J	268 (6.3%)
Nettle J	537 (12.7%)
Total number of speech episodes	4226

Speech episodes	Total (Percentage)
By female judge (3)	2339 (57.5%)
Kiefel CJ	1379 (33.9%)
Bell J	552 (13.6%)
Gordon J	408 (10.0%)
By male judge (4)	1728 (42.6%)
Gageler J	694 (17.1%)
Keane J	285 (7.0%)
Nettle J	462 (11.4%)
Edelman J	287 (7.1%)
Total number of speech episodes	4067

The slightly higher rates of male speech episodes in 2015–16, and of female speech episodes in 2017, can be explained by the ‘Chief Justice effect’, as both French CJ and Kiefel CJ represent the vast majority of speaking episodes. Putting that aside, these statistics deny any correlation between female judges’ volubility and their higher rates of interruption. This is demonstrated by the fact that the three female judges spoke for only 37% of the time in 2015–16, but still received more than half the total number of interruptions. Conversely, the slightly higher amount that female judges spoke in 2017 does not align with the far higher rate of interruption than that experienced by their male counterparts.

B Seniority

Putting aside gender, one might expect that junior judges would experience higher rates of interruption due to a greater degree of deference afforded to senior judges, as physically instantiated by their order of entrance and seating in the Court.[17] Despite the fact that a clear seniority model also applies in the US Supreme Court, Jacobi and Schweers found no evidence of a ‘freshman effect’,[18] and that seniority ‘[cuts] both ways’ in terms of interruptions between judges.[19] Similarly, I found a clear correlation between seniority and higher volubility, and that these two factors interacted to increase a judge’s chance of being interrupted. For example, disregarding the Chief Justice, the most senior judges (Kiefel J in 2015–16 and Bell J in 2017) both received the second highest number of interruptions, but also had among the highest numbers of speech episodes in both periods.[20] It is probable that senior judges are more active in oral argument due to their greater experience and confidence, and so may be more likely to be interrupted due to their higher level of engagement with advocates.

However, the explanatory power of seniority as a cause of interruptions is diminished by a comparison between judges of similar seniority and volubility but of different genders, such as Bell J and Gageler J. In 2015–16, these two judges each spoke for about 11% of the time. However, Bell J received 21% of the total interruptions, and Gageler J only 6%. This trend was amplified in 2017: Bell J spoke marginally more (14%) but received 26% of the total interruptions. On the other hand, the slight increase in interruptions of Gageler J (representing 15% of the total) in 2017 was proportionate to the 6% increase in his Honour’s speech episodes.

Figure 3: Volubility/Interruption Comparison of Gageler J and Bell J across 2015–16 and 2017

It is improbable that Bell J’s slight seniority over Gageler J[21] could explain the degree of discrepancy between their rates of interruption when considering that they spoke a similar amount in both periods. This trend is further borne out by a comparison of the most junior female judge with her male colleagues. Although the most junior male judge, Nettle J, spoke almost twice as much as Gordon J in 2015–16, his Honour was interrupted only slightly more than her Honour (9% compared with 7%). This trend was also evident in 2017, with the two judges speaking equivalent amounts of 10–11% but Gordon J receiving 13% of the total interruptions, which was 4% higher than the interruptions experienced by Nettle J.

The most pertinent comparison is that between Gordon J and Edelman J, given that the period of analysis includes their respective first years on the Court, in which they spoke the same amount.[22] Despite this, Edelman J represented only 3% of the interruptions in 2017, meaning that Gordon J was interrupted twice as much as his Honour in her Honour’s first year as a High Court Justice. These comparisons erode the salience of both seniority and volubility as explanations for these interruption patterns, and reinforce the finding of Jacobi and Schweers that, as more interaction effects are analysed, gender stands out as the primary cause.[23]

C ‘Female Register’

Alternatively, these gendered patterns may be explained by female judges speaking in a way that is more amenable to interruption. The notion of a distinct ‘female register’ has been a dominant theme in sociolinguistic literature since the 1970s, with studies finding that women’s speech tends to be hyper-polite and indirect, for example through use of ‘hedging’ language (eg ‘sort of’/‘probably’/‘kind of’) and declarative statements with rising intonation as opposed to direct questions.[24] These studies propose that women employ such language as an expression of powerlessness from their unequal social status.[25] By extension, Jacobi and Schweers argue that as interruptions often occur at the beginning of a judge’s question, the more frequent use of polite or indirect phrases (such as ‘sorry’, ‘could I ask’, or using an advocate’s name) by female judges delays the substance of their question and increases their likelihood of being interrupted.[26]

Although my study covered a much shorter period than that of Jacobi and Schweers, there were only a handful of instances in which a judge was interrupted after using polite or prefatory language.[27] More significantly, the use of such language was similar as between the female and male judges. In terms of general oral argument, Gageler J challenged the essentialist characterisation of the ‘female register’ by frequently using polite prefatory phrases and an advocate’s name before asking a question:

Yet, as noted above, his Honour was interrupted at much lower rates than his female colleagues, which casts doubt on a causal connection between ‘female speech’ and interruptions.

Jacobi and Schweers also found a clear downward trend in the use of polite prefatory phrases by female judges the longer they were on the Court, and proposed that the suppression of their ‘female register’ indicated an adaptation to masculine norms of speech in oral argument.[29] However, it is reasonable to infer that the speaking behaviour of US Supreme Court Justices is attenuated by the much shorter time frame in which they must question an advocate in comparison to High Court Justices.[30] Rather than a rejection of a putative ‘female register’, it may be that judges on the Supreme Court must simply adjust to the necessity of asking questions directly and quickly. Similarly, while female High Court Justices did not demonstrate the speech characteristics typically associated with their gender, this does not necessarily prove they are ‘honorary men’[31] who have imbibed the authority of the masculine register. It may also be explained by the fact that certain speaking patterns are adopted by all judges by virtue of their authoritative judicial role.[32] A similar speaking style should mean that all judges are interrupted at similar rates, and the fact that this is not the case brings the gendered pattern into sharper relief.

D Gendered Conversational Norms

If, then, we eliminate the judges’ own behaviour, the remaining explanation for these interruption patterns is that advocates perpetuate gendered norms during oral argument. Sociological research has found that, when conversing with women, men tend to speak more and for longer, monopolise the topic of conversation, and are more likely to interrupt or cut off their female partner.[33] Consequently, conversational interruptions can be read as an expression of dominance at the interpersonal level that reflects unequal gender power relations in society more broadly.[34] However, it would be reasonable to expect this norm to be suppressed in an elite setting such as a court, where power relations are mediated within the predetermined judge/advocate hierarchy.[35] Further, the formal structure and setting of oral argument is far removed from a casual interpersonal format, with the judges’ seniority[36] made explicit by their elevated seats and the Court’s hierarchised architecture.[37]

Nonetheless, my qualitative analysis revealed that female judges are subject to the same dominating behaviours in oral argument that they could expect in everyday conversation with men. I classified each interruption as falling into one of four categories:

1 Neutral: the advocate inadvertently speaks at the same time as a judge but stops once they have realised, or interrupts briefly to confirm a procedural query (eg page number, volume number of a law report);

2 Affirmatory: the advocate interjects to agree with the judge’s point or question;

3 Clarifying: the advocate interrupts to clarify a question or correct a factual point;

4 Pre-emptive: the advocate intentionally cuts off a judge’s speech by answering a question before it has been asked or diverts oral argument with their own point.

Self-evidently, each interruption differs in terms of its disruptive effect and in its signal as to the power relationship of the two speakers. The first category involves a simple miscommunication and the second can be characterised as a ‘back channel’ interruption which allows the interrupter to express encouragement and support of the speaker without cutting off their speech.[38] However, the fourth category can be characterised as an attempt to assert conversational dominance.[39] Not only were female judges interrupted more frequently, but the majority of those interruptions fell into this fourth category in both periods, representing 50% in 2017, and 46% in 2015–16. Male judges experienced a similar kind of pre-emptive disruption by having their speech intentionally cut off less frequently than their female colleagues, at an average rate of 30% over the two periods, keeping in mind that they also represented a far smaller amount of total interruptions.

I observed a range of different conversational behaviours in which a male advocate appeared to assert superiority during dialogue with a female judge — for example, by finishing a female judge’s sentence:

Secondly, male advocates would anticipate the gist of a question and answer before the female judge had finished asking it,[41] as illustrated in this exchange:

Pre-emptive interruptions were also employed defensively against female judges when advocates were challenged on their submissions:

In each of these instances, the judge lost the opportunity to ask her question after being interrupted, or had to wait for a lengthy diversion, by which time the oral argument had changed direction. This demonstrates what Sally Farley would classify as a successful interruption, as the floor has been completely ceded to the interrupter.[44] The fact that these intrusive interruptions disproportionately affected female judges indicates that such behaviour is not just zealous advocacy but an assertion of conversational dominance, which undermines the female judge’s capacity to ask meaningful questions of counsel. This indicates that the gendered behaviours observed in interpersonal dialogue more generally are replicated in oral argument in the High Court regardless of the institutionally assigned power of each speaker.

Moreover, the fact that the female judges experienced a comparatively higher rate of pre-emptive interruption also suggests that male counsel may have an unconscious discomfort in being corrected by female judges. Sociological research has found that a collocutor who assumes they occupy a higher position of power feels freer to interrupt a person of lower status, but expects criticism if the person they interrupt is of higher status.[45] Another study confirmed that interruptive behaviour therefore carries differential costs and rewards for men and women.[46] If men are socialised to use interruptions to assert conversational — and thus social — dominance, they may conversely feel that they lose this power when interrupted by a woman, which is generally ‘atypical’ gender behaviour.[47]

In this way, it is possible that male counsel, who are already socialised to feel more comfortable with interruptive behaviour, react to an incursion by a female judge on their assumed conversational dominance by pre-empting the correction, as demonstrated in the exchanges reproduced above. Yet this explanation would be more persuasive if the two speakers occupied equal positions of power. A study by Wiley and Woolley found that context affects the attribution of negative judgments in response to interruptive behaviour, whereby gender had a far less significant impact on producing negative judgements if the two speakers were equally powerful in a corporate context.[48] We might therefore expect that where women are in a more powerful position to their male collocutor, their interruptive behaviour would not be seen as an atypical gender behaviour that required correction through a pre-emptive attempt at regaining conversational dominance. The fact that this remains the case indicates the salience of gender dynamics over formal positions of power in determining conversational norms.

That even High Court Justices are not immune from unequal conversational norms underlines how powerful gender is as a relational framing device between individuals, even in the face of more relevant institutional hierarchies and rules of discourse.[49] This resonates with the disquieting realisation of Judith Resnik that ‘[i]f women judges do not escape the effects of being women, then no one can’.[50]

E Institutionalised Gender Bias

The different treatment of male and female judges by advocates also indicates that gender bias extends beyond the interpersonal to the institutional level on the High Court.

As discussed above, not all interruptions are inherently disruptive or dominating.[51] For example an ‘affirmatory’ interruption, through a linguistic ‘back channel’,[52] expresses an interrupter’s solidarity in order to build rapport with the speaker, which has a positive effect on their speech.[53] In a neat inversion of the category 4 statistics for female judges, 46% in 2015–16 and 50% in 2017 of the interruptions of male judges were brief, affirmatory interjections, such as:

Unlike a pre-emptive interruption, a judge is far more likely to be able to continue their point after an affirmatory interjection:

Of course, female judges also experienced affirmatory interruptions,[56] but less frequently than their male colleagues, which meant that a male judge had a better chance of being able to properly ask a substantive question of counsel.

More significantly, advocates demonstrated far greater deference to male judges after realising they had interrupted them. Although female judges were interrupted more often, there were fewer instances of this being acknowledged.[57] Even where an advocate did acknowledge an interruption, they often persisted, rather than ceding the floor back to the interrupted female judge, for example:

These reassertions of control over the conversational dynamic conform with the expectations of pre-emptive interruption discussed above, whereby interruptive behaviour by women may be perceived by male counsel as having a greater cost on their position as speakers and so prompts them to take preventative action.

In contrast, male advocates seemed far more cognisant of the cost of having interrupted a male judge. They would often cease speaking, or even ask permission to interrupt:

The horizontal bias demonstrated by these behaviours may be the product of an unconscious tendency to perceive men as more authoritative judicial figures.[62] According to Margaret Thornton, a long history of only appointing ‘Benchmark Men’ means that judicial merit is deeply gendered;[63] the values of reason, authority and merit in the Western legal tradition have been marked as ‘masculine, under the guise of universality’.[64] Meanwhile, the feminine ‘Other’ disrupts the male norm against which judicial difference is measured, as indicated by the qualifying adjective reserved for a ‘woman judge’.[65] Distrust of this ‘Other’ is expressed in conservative fears that female judges are appointed on the basis of their sex rather than merit,[66] and in applications for their recusal from cases involving ‘women’s issues’,[67] illustrating how any deviation from the natural position of the male judge is de-legitimised due to perceived ‘bias’.[68]

While women have now been ‘let in’ to even the highest positions of judicial power, it is questionable whether the legal discourses that constructed their authority have also changed.[69] This is evidenced by the propensity of advocates to show greater deference to male judges, whose approval they appear keener to attain. Conversely, the enduring association of the male archetype with judicial authority may explain why female judges experience conversational norms that treat them as though they are women first and judges second.

IV Normative Implications

A Women’s Judging and Participation

The interruption behaviour of advocates disproportionately affects female judges’ participation in the Court’s most important deliberative process. As has been observed in the US context, oral argument is an essential mechanism for judges to test arguments, refine key issues,[70] seek new information from advocates,[71] and, according to Roberts CJ, is an opportunity to crystallise ‘ideas that have been percolating’.[72] Given that Australian judges have much longer than 30 minutes[73] to discuss submissions with each side and that the High Court is thought to place greater emphasis on the role of oral argument,[74] the importance of this judge/advocate dialogue may be even more heightened in the Australian context. Accepting that this is a vital aspect of their decision-making, judges must have the opportunity to properly interrogate submissions and test hypotheticals with advocates during oral argument.

The fact that female judges were interrupted more often, and in a way that left their questions unanswered, means they are practically disadvantaged in their ability to pursue a line of thought in oral argument, which may impede informed decision-making processes and their ability to form coalitions with other judges.[75] This is particularly concerning given that the oral argument considered by this study occurred in Full Bench hearings, which concern the most important appellate and constitutional cases in the Court’s jurisdiction.[76] Consequently, while women clearly have formal equality of access to the Court, their lack of full participation in the institution’s most important deliberative process casts doubt over their substantive equality,[77] and in turn the Court’s integrity as a representative institution in a modern liberal society.[78]

B Is Representation Enough?

The ongoing disadvantages faced by women on the High Court Bench challenge feminist assumptions that a ‘critical mass’ of female representation necessarily improves the position of women in power. As reflected in United Nations strategies and the use of gender quotas around the world, there is a common belief that once female representation reaches 30% in an organisation, the ‘critical mass’ necessary for women to have a visible impact on the style and content of its decision-making is reached.[79] Yet my study revealed that an almost equal representation of female judges, and even a female Chief Justice, is not enough to overcome subtle gender discrimination.

Of course, there are many reasons why the presence of women on the High Court should be encouraged for symbolic reasons: it signals an equality of opportunity for women in the legal profession; encourages other women to aspire to higher judicial appointment;[80] and disrupts the ‘aesthetic of the [male] judicial norm’.[81] On a more practical level, Rosemary Hunter argues that women have an educative effect on male colleagues by not allowing sexist comments and gender bias to ‘go unquestioned’.[82] This civilising influence was demonstrated by Kiefel J in at least one exchange in the transcripts:

While such corrections are laudable, it is disappointing that as Chief Justice, Kiefel CJ was unable to similarly prevent the male advocates interrupting her female colleagues at much higher rates. The fact that interruptions of female judges increased under her Honour’s leadership may be due to a fear of aligning herself with a ‘woman’s issue’ and deviating from the (masculine) norm of impartiality, which would be inimical to the performance of neutrality necessary for the judicial role.[84] We need only look at the accusations of bias against Supreme Court Justice Sonia Sotomayor when she identified her judicial personality as a ‘wise Latina woman’[85] to understand the harm done to a female judge’s legitimacy when she claims a ‘different voice’.[86] In a similar way, Kiefel CJ could undermine her own authority if she was perceived as a ‘feminist judge’ (assuming that she was conscious of the differential interruptions), especially as female judges tend to demonstrate neutrality beyond what is expected of male judges in order to eschew any appearance of ‘difference’.[87] This may explain why female judges on the High Court have been called a ‘particular disappointment’ in the media, as they are seen to reproduce the ‘old styles of the law’,[88] rather than bringing fresh new ‘feminine’ perspectives to judging.[89] Putting aside the essentialist contention that women judge differently,[90] this sense of disappointment reflects the unreasonable expectation of liberal feminism that women could transform institutions simply by their presence.[91] Instead, my study suggests that substantive equality in the High Court can only be achieved by tackling the patriarchal norms and discourse that are embedded in society and replicated in its institutions.[92]

V Solutions

A Chief Justice Regulation

The challenges that would face a female Chief Justice in addressing gendered interruption behaviour means the most pragmatic solution is to equip male judges as agents of change. In their respective studies on gendered interruption in the US Supreme Court, Jacobi and Schweers,[93] as well as Feldman and Gill,[94] conclude that the Chief Justice should take on greater responsibility in enforcing the ‘turn-taking’ norm and preventing the frequent interruption of female judges. As noted by Justice Bertha Wilson, change in the legal profession is more likely to arise when it is championed by ‘impartial’ groups such as professional bodies rather than by so-called ‘special interests’ such as women’s advocacy groups.[95] Such an ‘impartial advocate’ with the necessary authority to institute change could be a male Chief Justice. Indeed, Beverley Baines found that male judges in Canada use the term ‘feminist’ more frequently in their judgments than female judges,[96] which is probably because their position of unquestioned neutrality insulates them from accusations of bias. This may be one reason why interruptions of female judges were lower under French CJ than Kiefel CJ, along with his Honour’s longer experience as Chief Justice, and possibly greater confidence to intercede and enforce turn-taking in oral argument:

Consequently, any future male Chief Justice should be made aware of the stark disparity in the rates of interruption between male and female judges so that they can consciously mitigate these trends when regulating oral argument. As women may be seen as ‘biased’ for defending fellow female judges against gendered interruption, a male judge may carry more widespread legitimacy in preventing this behaviour in the courtroom and encouraging institutional change.[98]

B Training

In terms of addressing the underlying problem, male advocates must be made aware of their unconscious bias. Some psychological studies on racial prejudice have found that individuals are less likely to exhibit implicit biases when they are placed in a subordinate position to a person from a stigmatised group.[99] However, given that male advocates interrupted female judges even more frequently under a female Chief Justice, my study indicates that similar situational exposure to women in power is not sufficient to overcome implicit gender bias. Rather, specific training on unconscious bias is necessary. Other studies have found that long-term reductions in implicit racial biases were achieved simply by training an individual to be aware of their habitual modes of thinking, and exposing them to conscious, non-prejudiced attitudes.[100] In a similar way, as male advocates most likely do not consciously interrupt female judges at a higher rate, this behaviour could be diminished simply by informing them of this tendency and lodging this ‘artifact of conversation’[101] in their mind when appearing in argument before the High Court, or indeed in any conversation. Law schools might similarly have a role in educating students on how unconscious biases are reproduced in the courtroom and in legal culture more broadly.

VI Conclusion

While women have come a long way in the legal profession and now lead the High Court, this article has shown how it is not sufficient to simply ‘[add] women ... and [stir]’,[102] and expect them to achieve substantive equality in a male-dominated institution. In particular, this empirical study gives us cause to reassess the progress that appears to have been achieved in the legal profession, particularly by the appointment of a female Chief Justice. It shows that having women in superior positions of institutional authority is insufficient to transcend the fact that High Court dialogue is embedded in broader social discourses and power relations, meaning that female judges are treated as conversational inferiors and denied full participation in oral argument. Consequently, this article should draw attention to the fact that despite women’s formal equality, they still face barriers in being treated as equal to their male colleagues, even at the pinnacle of their legal careers. To address this issue, we must consciously draw male advocates’ attention to their differential treatment of High Court Justices.

More importantly, the unequal treatment of men and women even at the height of the legal profession should be a wake-up call to the need for systemic change in the ontology of legal authority. In particular, the comparatively higher levels of deference shown by predominantly male counsel towards male judges confirms that masculinity is the ‘natural, legitimate property of the judicial field’.[103] While Sommerlad notes that women’s greater presence as judges has begun to challenge the ‘homologous relationship between masculinity and legal authority’,[104] this cultural change is constrained by the law’s refusal to ‘recognise’ different judicial identity claims under the guise of impartiality, which reproduces masculine cultural hegemony.[105] In this sense, meaningful cultural change in the profession will rely not just on recognising the reproduction of conversational bias in the courtroom, but in dismantling the specifically masculine discourses and power structures that continue to define judicial merit and result in female judges being treated as women first and judges second.

The next frontier of this research would be to verify the findings of this study over a larger scale and time period, in order to confirm whether these patterns can be detected since Gaudron J was appointed as the first female High Court Justice in 1987. It would also be valuable to interview the female judges in order to determine whether they are aware of interruption behaviours,[106] and to what extent it materially impacts on the task of judging.

VII Appendix I: Statistics Organised by Interruptee

2015–16 (n = 22)	Total (Percentage)
Total interruptions of female judge	115 (52.0%)
By male counsel	113 (51.1%)
By female counsel	1 (0.45%)
By male judge	0
By female judge	1 (0.45%)
Total interruptions of male judge	106 (47.8%)
By male counsel	101 (45.7%)
By female counsel	3 (1.4%)
By male judge	0
By female judge	2 (0.9%)
Total interruptions	221

2017 (n = 23)	Total (Percentage)
Total interruptions of female judge	136 (69.0%)
By male counsel	123 (62.4%)
By female counsel	12 (6%)
By male judge	1 (0.5%)
By female judge	0
Total interruptions of male judge	61 (30.96%)
By male counsel	54 (27.4%)
By female counsel	7 (3.5%)
By male judge	0
By female judge	0
Total interruptions	197

VIII Appendix II: Full List of Transcripts[107]

A 2015 Full Bench Transcripts

Plaintiff M68/2015 v Minister for Immigration and Border Protection [2015] HCATrans 255

Plaintiff M68/2015 v Minister for Immigration and Border Protection [2015] HCATrans 256

Commonwealth v Director, Fair Work Building Industry Inspectorate [2015] HCATrans 259

B 2016 Full Bench Transcripts

New South Wales Aboriginal Land Council v Minister Administering the Crown Lands Act [2016] HCATrans 228

New South Wales Aboriginal Land Council v Minister Administering the Crown Lands Act [2016] HCATrans 229

C 2017 Full Bench Transcripts

Aldi Foods Pty Ltd v Shop, Distributive & Allied Employees Association [2017] HCATrans 149

Probuild Constructions Pty Ltd v Maxcon Constructions Pty Ltd [2017] HCATrans 226

* BA, LLB (Hons [1]) (UNSW). I thank Professor Rosalind Dixon for her ongoing advice and encouragement in the development of this article. I also acknowledge the anonymous reviewers for their helpful comments and the members of the Review for their excellent editorial work. All errors remain my own.

[1] Transcript of Proceedings, Bell Group NV (in liq) v Western Australia [2016] HCATrans 79, 6143–9 (‘Bell Group (Transcript)’).

[2] Marianne LaFrance, ‘Gender and Interruptions: Individual Infraction or Violation of the Social Order?’ (1992) 16(4) Psychology of Women Quarterly 497, 497.

[3] Emanuel A Schegloff, ‘Overlapping Talk and the Organization of Turn-Taking for Conversation’ (2000) 29(1) Language in Society 1.

[4] LaFrance (n 2) 498–9; Hiroko Itakura, ‘Describing Conversational Dominance’ (2001) 33(12) Journal of Pragmatics 1867–8.

[5] Don H Zimmerman and Candace West, ‘Sex Roles, Interruptions and Silences in Conversation’ in Barrie Thorne and Nancy Henley (eds), Language and Sex: Difference and Dominance (Newbury House Publishers, 1975) 105, 116–17.

[7] Cathryn Johnson, ‘Gender, Legitimate Authority, and Leader-Subordinate Conversations’ (1994) 59(1) American Sociological Review 122, 133. See generally Nina Eliasoph, ‘Politeness, Power, and Women’s Language: Rethinking Study in Language and Gender’ (1987) 32 Berkeley Journal of Sociology 79, 80–1.

[8] Tonja Jacobi and Dylan Schweers, ‘Justice, Interrupted: The Effect of Gender, Ideology, and Seniority at Supreme Court Oral Arguments’ (2017) 103(7) Virginia Law Review 1379, 1493. I gratefully acknowledge the advice of Professor Jacobi in developing the research methodology for this study.

[9] Interruptions by male advocates represented 96.8% of interruptions in 2015–16 and 90% in 2017. The number of female advocates was so small that their interruption behaviour was statistically insignificant, and there was only one case in the three years in which both sides were represented by female counsel: Transcript of Proceedings, IL v The Queen [2017] HCATrans 65, in which Belinda Rigg SC and Sally Dowling SC appeared. For a full breakdown of the effect of interuptees’ gender, see below Part VII. These statistics reflect the broader under-representation of women at the higher levels of the legal profession: Daniel Reynolds and George Williams, ‘Gender Equality among Barristers before the High Court’ (2017) 91(6) Australian Law Journal 483.

[11] I verified that this symbol signified interruptions by reviewing an AV recording of a Full Bench hearing, in which the ‘- - -’ corresponded with interruptions of dialogue in the proceedings. I counted each interruption by hand and then verified my findings three times when reanalysing the transcripts for the purpose of qualitative analysis. I also counted instances where ‘- -’ was erroneously used to indicate an interruption.

[12] Following the lead of Adam Feldman and Rebecca Gill, ‘Echoes from a Gendered Court: Examining the Justices’ Interactions during Supreme Court Oral Arguments’ (Working Paper, 31 January 2017) <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2906136>, archived at <https://perma.cc/A6D8-6JQ3>. The differential effects caused by different kinds of interruptions in the context of oral argument is analysed in Part III.

[13] Where each individual total is rounded to the nearest whole number, as are all totals discussed in this part. Note that in the final Full Bench hearing for this period (Perara-Cathcart v The Queen [2016] HCATrans 269), there were only six judges and an equal number of male and female judges as French CJ had retired and Edelman J had not begun his appointment.

[14] Chief Justice Roberts of the US Supreme Court likened the position to that of an ‘umpire’: Jacobi and Schweers (n 8) 1495–6.

[16] A ‘speech episode’ represents each judge’s uninterrupted period of dialogue, following the approach of Jacobi and Schweers (n 8) 1435, 1438–9.

[18] Jacobi and Schweers (n 8) 1430–1. A possible ‘freshman effect’ was specifically controlled as a variable in the study of Feldman and Gill (n 12) 52.

[20] Bell J represented 14% of the speech episodes in 2017, compared to Gageler J’s 17%, though his Honour also received the highest rate of male judge interruptions that year.

[21] Bell J having been appointed to the Court in 2009, three years before Gageler J: ‘About the Justices’, High Court of Australia (Web Page) <http://www.hcourt.gov.au/justices/about-the-justices> , archived at <https://www.perma.cc/A53F-2M2F>.

[22] Both representing around 7% of the total number of speech episodes in 2015–16 and 2017 respectively.

[23] Jacobi and Schweers (n 8) found that by excluding gender in their regression analysis, ‘the significance of the effect of seniority drops away in almost every model’: at 1464. They infer that seniority therefore only ‘seems significant because gender is so powerful’, ie that gendered interruption behaviour is ‘masked by the norm of seniority’.

[24] Faye Crosby and Linda Nyquist, ‘The Female Register: An Empirical Study of Lakoff’s Hypotheses’ (1977) 6(3) Language in Society 313, 313–14; Janet E Ainsworth, ‘In a Different Register: The Pragmatics of Powerlessness in Police Interrogation’ (1993) 103(2) Yale Law Journal 259, 261–3; Adrienne B Hancock and Benjamin A Rubin, ‘Influence of Communication Partner’s Gender on Language’ (2015) 34(1) Journal of Language and Social Psychology 46, 47.

[27] I observed six examples from male judges and seven from female judges in 2015–16, and five from male judges and six from female judges in 2017.

[28] Transcript of Proceedings, PT Bayan Resources TBK v BCBC Singapore Pte Ltd [2015] HCATrans 184, 1784 (‘PT Bayan (Transcript)’).

[30] Supreme Court Justices usually only have 30 minutes to hear one side of a case, as opposed to single- or multiple-day hearings on the High Court: Eve M Ringsmuth, Amanda C Bryan and Timothy R Johnson, ‘Voting Fluidity and Oral Argument on the US Supreme Court’ (2013) 66(2) Political Research Quarterly 429, 432.

[32] Of course, this may itself be masculinised, which is discussed further in Part III(E) below.

[33] Anthony Mulac, ‘Men’s and Women’s Talk in Same-Gender and Mixed-Gender Dyads: Power or Polemic?’ (1989) 8(3–4) Journal of Language and Social Psychology 249, 259–63.

[36] This seniority can be expected to attract deference despite the fact that the High Court Rules 2004 (Cth) do not explicitly preclude interruptions of judges, unlike the US Supreme Court equivalent: Supreme Court of the United States, Guide for Counsel in Cases to Be Argued before the Supreme Court of the United States (October 2017) 9, archived at <https://perma.cc/CHC6-43UV>, which directs advocates to ‘[n]ever interrupt a Justice who is addressing you ... If you are speaking and a Justice interrupts you, cease talking immediately and listen’.

[37] It is well recognised that courtrooms are structurally designed to define relationships of power between participants in the space: see, eg, Linda Mulcahy, Legal Architecture: Justice, Due Process and the Place of Law (Routledge, 2011) 14–15, 38.

[39] See generally Penelope Brown and Stephen C Levinson, Politeness: Some Universals in Language Usage (Cambridge University Press, 1987) 245–9, discussing how social strata are reflected in conversational patterns.

[41] Very similar behaviour was also observed in male judges (rather than just advocates) by Jacobi and Schweers (n 8) 1413.

[44] Sally D Farley, ‘Attaining Status at the Expense of Likeability: Pilfering Power through Conversational Interruptions’ (2008) 32(4) Journal of Nonverbal Behavior 241, 252.

[46] Mary Glenn Wiley and Dale E Woolley, ‘Interruptions among Equals: Power Plays That Fail’ (1988) 2(1) Gender and Society 90, 98–9.

[49] Cecilia L Ridgeway, ‘Gender as a Group Process: Implications for the Persistence of Inequality’ in Shelley J Correll (ed), Social Psychology of Gender (Elsevier, 2007) 311, 312–18.

[50] Judith Resnik, ‘Asking about Gender in Courts’ (1996) 21(4) Signs 952, 971–2.

[51] Deborah Tannen, Conversational Style: Analyzing Talk among Friends (Cambridge University Press, 1984). For example, positive interruptions may actually encourage people to speak more and dominate talking-turns: at 118–24.

[53] Tali Mendelberg, Christopher F Karpowitz and J Baxter Oliphant, ‘Gender Inequality in Deliberation: Unpacking the Black Box of Interaction’ (2014) 12(1) Perspectives on Politics 18, 21.

[54] Transcript of Proceedings, Probuild Constructions (Aust) Pty Ltd v Maxcon Constructions Pty Ltd [2017] HCATrans 226, 3708–11.

[55] Transcript of Proceedings, Graham v Minister for Immigration and Border Protection [2017] HCATrans 63, 1063–8.

[56] This category represented 19% of the interruptions of female judges in 2015–16 and 25% in 2017.

[57] This conforms with the finding of Jacobi and Schweers (n 8) that male judges and advocates were the least likely to recognise when they had interrupted a woman despite the frequency of this occurrence: at 1469.

[59] Transcript of Proceedings, Day v Australian Electoral Officer (SA) [2016] HCATrans 98, 2358–64.

[62] See Rosalind Dixon, ‘Female Justices, Feminism, and the Politics of Judicial Appointment: A Re-Examination’ (2010) 21(2) Yale Journal of Law and Feminism 297, 334–5, arguing that male behaviours are associated with legal talent and merit.

[63] Margaret Thornton, ‘“Otherness” on the Bench: How Merit Is Gendered’ [2007] SydLawRw 16; (2007) 29(3) Sydney Law Review 391, 404. Thornton identifies the ‘Benchmark’ as ‘white, heterosexual, able-bodied, middle class men’: at 394.

[65] Regina Graycar, ‘The Gender of Judgments: Some Reflections on Bias’ (1998) 32(1) University of British Columbia Law Review 1, 2–3 (‘The Gender of Judgments’). See also Ulrike Schultz and Gisela Shaw, ‘Introduction: Gender and Judging’ in Ulrike Schultz and Gisela Shaw (eds), Gender and Judging (Hart Publishing, 2013) 3, 24, where they describe the traditional prototype of the judge as an ‘outstanding white male bourgeois personality of high morality, humanistic, mild, the loving father of a happy family ... covering their individuality with a black robe’.

[69] Rosemary Hunter, ‘More than Just a Different Face? Judicial Diversity and Decision-Making’ (2015) 68(1) Current Legal Problems 119, 127–8 (‘More than a Different Face’). See also Graycar, ‘The Gender of Judgments’ (n 65) 20; Hilary Sommerlad, ‘Let History Judge? Gender, Race, Class and Performative Identity: A Study of Women Judges in England and Wales’ in Ulrike Schultz and Gisela Shaw (eds), Gender and Judging (Hart Publishing, 2013) 355, 358.

[70] Barry Sullivan and Megan Canty, ‘Interruptions in Search of a Purpose: Oral Argument in the Supreme Court, October Terms 1958–60 and 2010–12’ [2015] (5) Utah Law Review 1005, 1024–5, cited in Jacobi and Schweers (n 8) 1394.

[71] Timothy R Johnson, Oral Arguments and Decision Making on the United States Supreme Court (State University of New York Press, 2004) 55, cited in Jacobi and Schweers (n 8) 1394.

[74] Russell Smyth and Vinod Mishra, ‘Barrister Gender and Litigant Success in the High Court of Australia’ (2014) 49(1) Australian Journal of Political Science 1, 2.

[77] Kate Malleson, ‘Justifying Gender Equality on the Bench: Why Difference Won’t Do’ (2003) 11(1) Feminist Legal Studies 1, 17, discussing substantive equality in the context of female appointments to courts as part of an affirmative action strategy as accommodating ‘more sophisticated interpretations of merit and competence’.

[79] Report of the Fourth World Conference on Women, UN Doc A/CONF.177/20/Rev.1 (1996) 205; Mendelberg, Karpowitz and Oliphant (n 53) 18.

[83] Transcript of Proceedings, Crown Melbourne Ltd v Cosmopolitan Hotel (Vic) Pty Ltd [2016] HCATrans 103, 3291–302.

[89] McLoughlin, ‘A Particular Disappointment?’ (n 88) 280. McLoughlin interprets Waterford’s (n 88) reference to the hoped-for ‘wider perspectives’ from female judicial appointments as an expectation that they would employ ‘feminine methods of [judicial] reasoning’: at 274, 280.

[90] See, eg, Chief Judge Mary M Schroeder, ‘Judging with a Difference’ (2002) 14(2) Yale Journal of Law and Feminism 255. Schultz and Shaw (n 65) identify a number of areas of law in which female judges reached different outcomes to their male counterparts or judged with a different approach: at 34–7. See generally Rosemary Hunter and Danielle Tyson, ‘Justice Betty King: A Study of Feminist Judging in Action’ [2017] UNSWLawJl 28; (2017) 40(2) University of New South Wales Law Journal 778, in which the authors note the potential for ‘different judging’ from a feminist perspective.

[96] Beverley Baines, ‘Must Feminist Judges Self-Identify as Feminists?’ in Ulrike Schultz and Gisela Shaw (eds), Gender and Judging (Hart Publishing, 2013) 379, 388.

[98] This practical reality was recognised by Justice Ruth Bader Ginsburg when she argued that a male Justice is more likely to be heard on gender issues than ‘pro feminist’ female Justices: Joan Biskupic, ‘Ginsburg: Court Needs Another Woman’, ABC News (Web Page, 5 May 2009) <https://abcnews.go.com/Politics/ginsburg-court-woman/story?id=7513795>, archived at <https://perma.cc/DA2Z-6Z26>.

[99] Brian S Lowery and Curtis D Harden, ‘Social Influence Effects on Automatic Racial Prejudice’ (2001) 81(5) Journal of Personality and Social Psychology 842.

[100] Patricia G Devine et al, ‘Long-Term Reduction in Implicit Race Bias: A Prejudice Habit-Breaking Intervention’ (2012) 48(6) Journal of Experimental Social Psychology 1267; Karen Gonsalkorale et al, ‘Mechanisms of Group Membership and Exemplar Exposure Effects on Implicit Attitudes’ (2010) 41(3) Social Psychology 158.

[107] Note that for the purpose of the study, multiple-day hearings were counted as one case.

Melbourne University Law Review