Steven Price

My book

Media Minefield

Guide to NZ Media Law

Official Information Act

Official Information Act

Bill of Rights Act

Media law resources

Feeds (RSS)


Evidence about the chilling effect of ordering journalists to burn confidential sources

19 June 2009

  My background and expertise  

  1. I am a barrister specialising in media law. I am also an adjunct lecturer at Victoria University of Wellington, where I have been teaching media law at graduate and post-graduate level for about 10 years.


  1. I am also the author of Media Minefield: A Journalists’ Guide to Media Regulation in New Zealand, a handbook for journalists which provides guidance on, among other things, legal and ethical issues relating to source confidentiality.


  1. I have a law degree with first class honours from Victoria University of Wellington. I also have a Master of Journalism degree from the Graduate School of Journalism at the University of California at Berkeley, where I studied on a Fulbright scholarship from 1996-1998.


  1. Between 1993 and 1996 I worked at Kensington Swan as a staff solicitor and spent much of that time working with the TVNZ consumer affairs programme Fair Go as a client, on a range of media law issues.


  1. I also have some experience as a working journalist. I was a general assignment reporter for the Hot Springs Sentinel Record in Arkansas for a year following my graduation from Berkeley. I also completed short internships at the Oakland Tribune, KQED radio, WorldNet Television and Voice of America.


  1. In New Zealand after 1999, I wrote a column on legal issues for several years for the Evening Post, Otago Daily Times and NZ Herald, and had freelance work published in a range of media outlets, including the Listener, the National Business Review, Metro, North and South, the Dominion, and the Christchurch Press. I have also worked for Radio New Zealand programmes, producing a series on legal issues, providing regular commentary on legal and media matters on Nine to Noon and reporting for MediaWatch.


  1. I have experience with source confidentiality issues across a range of roles. I have taught students about the issue. I have been involved in ethics training touching on this issue for Radio New Zealand and Fairfax. I have supervised student papers on it. I have researched and reported on the issue for MediaWatch. I have dealt with confidential sources myself in the course of my own reporting. I have been a source myself, occasionally speaking to other reporters in confidence. I have provided legal advice to other journalists, notably Nicky Hager, who were dealing with confidential sources. I have argued a case in the Court of Appeal where an author was accused of breaching confidences.


  1. In the course of preparing this affidavit I have spend most of a week researching the available empirical evidence on the question of the chilling effect of court-ordered disclosures of confidential sources on the flow of important information to the public via the media. (I list at the end of this affidavit the sources I have consulted. I note that the academic commentary on the chill effect is voluminous, particularly in the United States, and there is much that I have not read).


  1. I have read the Code of Conduct for expert witnesses and agree to comply with it.

 The issue   

  1. I have read the affidavits of Gavin Peter Ellis, John James Campbell, Carol Ann Hirschfeld, Ingrid Marieke Leary and James Mark Jennings, filed in this proceeding.


  1. I understand that it is my role to comment on the suggestion, best expressed in Mr Ellis’s affidavit, that: “There will be a chilling effect if informants harbour doubts about journalists’ willingness or ability to protect identity.” In particular, it is suggested that the flow of publicly important information to journalists from such sources will be impeded if the courts do not protect their promises of confidentiality (this is known as “the chilling effect”).


  1. I believe this issue falls within my area of expertise.

 The short answer  

  1. It is impossible to say for sure whether there is a chilling effect, or if there is, how great its impact may be. There is no direct and compelling empirical evidence about it, anywhere. This should not be surprising. There is no easy way to tell why sources come forward, how they would have behaved if the risks had been different, and how they are likely to behave in the future. If sources don’t come forward, we don’t find out, and we can’t know why they didn’t. Indeed there is no reason to assume that they all think the same about the legal risks, if they think about them at all.


  1. Such empirical evidence as there is tends to be based on the assertions and surveys of journalists. The US Supreme Court noted in the leading case of Branzburg v Hayes that such evidence is “speculative” and must be viewed in the light of journalists’ “professional self-interest”.


  1. I would note that other privileges, such as legal professional privilege and medical privilege, are not based on empirical evidence either. The harm there is regarded as self-evident.

 The longer answer  

  1. It seems very likely that there is some sort of chilling effect. Almost all judges, academics, law reform agencies, and legislatures that have considered this issue, believe that there is. It follows as a matter of logic that sources concerned about keeping their identities secret may think twice before providing information to journalists who may be compelled to reveal them. Most countries have some form of protection for journalists with respect to their confidential sources for this reason.


  1. There is ample evidence of very significant stories that could not have been written without the help of confidential sources, including some of the most publicly important journalism of our time. In the United States, for example, those stories include Watergate, the My Lai massacre, the Pentagon Papers, the Iran-Contra scandal, Abu Ghraib, and the federal government’s involvement in unauthorised wire-tapping and torture.


  1. However, the chilling effect is not straightforward. The crucial issue is whether (and how many) important confidential sources would fail to come forward if they perceived some possibility that their identities would be revealed by journalists in the course of future litigation. I am inclined to think that some journalists’ statements predicting the “drying up” of sources are exaggerated. Indeed that seems clear when we recall that very few jurisdictions provide journalists with an absolute privilege to protect their promises of confidentiality. The fact that journalists can be (and occasionally are) compelled to reveal their sources (and occasionally do) has not deterred such investigative journalism as has occurred – including the stories listed above from the United States, where there is no federal statutory journalists’ privilege.


  1. Similarly, in the United Kingdom there have been several highly publicised court rulings that have not supported journalistic confidentiality, but there is still a vibrant press containing many stories based on confidential sources, such as the recent MPs’ expenses scandal in the United Kingdom, which came from a leak that was probably illegal. In Australia and Canada, too, the protection for journalists’ sources is limited. So it is not clear that the legal possibility that some journalists will be ordered to disclose confidential sources has had any effect on investigative journalism based on confidential sources.


  1. On the other hand, it should be noted that many court decisions have supported journalists’ promises of confidentiality. For example, in the United States, some federal court decisions have applied a common law source protection privilege, and in the United Kingdom there is at least a measure of statutory protection in the Contempt of Court Act, even if it is not often successfully invoked. In addition, when those arguments fail, journalists will often go to jail rather than reveal their sources. Both these factors may bolster sources’ confidence in the media’s ability to protect them, even bearing in mind possible court applications.


  1. The following observations are, I think, relevant to assessing the likely chilling effect:

 Obtaining confidential information 

  1. Such empirical evidence as there is suggests that journalists use confidential sources in a quarter to a third of stories. (Blasi, 1972; Osborne, 1985). They are of more significance for investigative and beat reporters than for journalists working in high turnover newsrooms, such as wire services.


  1. Those sources are used for a variety of reasons, including verifying material independently obtained, providing expert context or partisan analysis, identifying other sources, determining which stories to pursue and with what resources, assessing the emphasis to give a certain fact. This is very helpful to journalists and ultimately the public, but it is only on rare occasions that such sources themselves provide explosive information.


  1. Blasi finds that journalists seldom discuss the terms of their confidentiality with sources. Some journalists deliberately avoid mentioning the possibility. While sometimes express promises of confidentiality are made, the journalist and source will seldom discuss the extent to which the law may undermine those promises or the circumstances in which the journalist may be prepared to breach it. Many newsrooms do not have comprehensive policies about the use of confidential sources. All of this accords with my knowledge and experience of the modern situation in New Zealand.


  1. It seems very likely that confidential sources are almost always primarily interested in keeping their name out of the media in the ensuing coverage. Their other main concern is usually whether the journalist is able to understand their information and report it accurately (Blasi 1972). It is likely that few think about the possibility of a future lawsuit in which their identities may be relevant.


  1. In part, this is probably because it is rare for subpoenas to seek journalists’ confidential sources. United States evidence suggests that when subpoenas are issued against the media, they very rarely seek the revelation of confidential sources, as opposed to eye-witness evidence, information about non-confidential sources and other information (Alexander, Perry and Chamberlain 1994). That survey found that 726 subpoenas were issued against news organisations in Florida between 1988 and 1992, but fewer than 3% were for confidential sources and “Most respondents said the threat or use of subpoenas against their news organizations has not affected their policy on use of confidential sources or on the retention of notes, drafts and similar materials.”


  1. In any event, the incidence of subpoenas for any sort of journalists’ evidence is extremely low in New Zealand. In part, that is because the media themselves will not generally reveal their sources in a civil lawsuit and the police are usually loath to formally involve the media in a criminal one. The police attitude may have been influenced by the guidelines developed by the courts in relation to searches of media premises in TVNZ v Attorney-General [1995] 2 NZLR 641. It may also be influenced by the fact that, as Michael Kinsley puts it, “the press are uniquely able to make their screams of pain heard.” Police use of their subpoena powers against the media is likely to provoke a strong and well-publicised backlash. It may also be because journalists seldom possess information that is strongly useful as evidence.


  1. Because there have been few such cases in New Zealand, it is possible that one big case attracting much media attention, as here, may have a disproportionate impact on the attitudes of sources. It is also possible that in future there will be greater use of subpoena powers against the media.

  Breaching the confidence 

  1. Ethically, there is great debate about the circumstances in which a journalist is entitled to breach a confidence. I note that there are even inconsistencies between the journalists who have given affidavits in this case. I have found that most journalists agree that they may breach a promise of confidentiality to save lives. Some would breach a confidence to prevent an innocent person being convicted. Some would voluntarily provide the authorities with confidential information to help secure a conviction for a significant crime (see Blasi). Some would say the promise no longer needs to be honoured if the source deliberately misled the journalist. Some have even done so because they regard breaching the confidence as being in the public interest. For example, on the eve of an election, a reporter received information from a Republican source damaging to a Democratic candidate. It was subject to a promise of confidentiality. The newspaper formed the view that the better story was that the Republican source was trying to peddle damaging information and published the name of the source. (Cohen v Cowles Media). Kase (1989) points to many examples of journalists voluntarily exposing confidential sources.


  1. There have been similar examples in New Zealand. See, for example: Hood v TVNZ (where a TVNZ journalist broke a verbal and written undertaking to a source not to film or quote him) and Diocese of Dunedin et al v TV3 (where TV3 broadcast footage of an interviewee it had secretly filmed after promising not to identify him). Both led to criticism from the Broadcasting Standards Authority. In my view, such episodes are more likely to produce a chilling effect on other potential sources than a court ordered disclosure. However, there does not seem to be any evidence of any such chilling effect. It is probably relevant that these episodes were not widely publicised.


  1. It is clear that many journalists and media organisations strongly believe that they and not the courts should decide in what circumstances they should be entitled to breach their confidences.

 The evidence of the chilling effect  

  1. There is overseas evidence from journalists that the possibility that they will have to reveal sources has hurt their reporting. Blasi finds evidence of journalists describing particular valuable sources who have gone to ground because of a fear of subpoenas. 8% of journalists he surveyed said their coverage in the past 18 months had been adversely affected by the possibility that they may be subpoenaed, and another 11% said they were not sure. Overall, however, Blasi concludes that “the adverse effect of the subpoena threat has been primarily in “poisoning the atmosphere” so as to make insightful, interpretative reporting more difficult, rather than causing sources to “dry up” completely”.


  1. In 1985, Osborne found that  two thirds of Pulitzer Prize nominees felt that confidential sources had played a significant role in their stories – but only 18% felt their coverage generally had been adversely affected by the possibility they would have to disclose sources (another 20% were unsure). Osborne’s survey response rate, however, was only 30%. He concluded that “the steady issuance of press subpoenas and the jailing of reporters in a number of celebrated cases have apparently failed to deter the majority of journalists from maintaining confidential source relationships.”


  1. One study found that newspapers in American states with journalist-protecting shield laws do more investigative reporting and win more awards for their reporting than those in states without such laws (Wirth 1995).


  1. In Australia, new anti-terrorism laws have provided sweeping new powers to the government to access journalists’ confidential materials, but a recent survey of Australian journalists and their lawyers found “absolutely no evidence of reluctance to pursue stories that might place journalists in a difficult position with respect to the need to protect their sources in the face of prosecution.” (McNamara 2009).


  1. In a highly publicised 2005 incident, the editor of the Cleveland Plain Dealer announced that the paper had withheld two significant stories from publication because they were based on illegally leaked documents and could lead to penalties against the paper and the journalists. (It does seem, however, that their sources were not deterred, and other editors have reportedly said they were baffled by this decision). See McFadden 2005 and Johnston 2005.


  1. On the other hand, there are journalists who doubt the existence of a significant chilling effect.  For example, Walter Pincus, an investigative reporter for the Washington Post who was subpoenaed in two very high profile recent cases, told the New Yorker that he was sceptical of the chilling effect and that “my sources are not drying up… it hasn’t hurt me.”


  1. Some news organisations such as the Dallas Morning News have developed a policy of warning confidential sources that in rare instances the journalist may be forced to identify them if efforts to protect them are exhausted in court. There is no indication that the sources for these organisations have dried up. (Eliason, 2007)

 To leak or not to leak? 

  1. In the absence of hard data about sources’ behaviour, we can only speculate about what goes on in their minds. It seems reasonable to observe that many confidential sources provide information that is not remotely likely to end up being relevant to a court case. Others will perceive that the risk of such a court case is very small. Neither type of source is likely to be deterred by an occasional disclosure order, particularly if it relates to circumstances far removed from those in which they are providing information.


  1. On the other hand, when significant information is provided confidentially to the media – the sort of information that leads to stories of significant public interest – it will often involve a breach of confidence and possibly a breach of employment obligations on the part of the informant. Significant stories also involve a greater than usual chance of a defamation action. Some leaks themselves involve breaches of the criminal law (such as the provision of classified information or stolen documents). Others (as here) involve confessions of crimes.


  1. Thus, in some circumstances, the possibility of the source’s identity coming out in a court case is more likely to be present to the source’s mind. The chilling effect may be strongest here. However, it will be one factor among many: the importance to the source of the public learning the information; the likelihood of being discovered through other means (such as a leak inquiry or a slip-up by the journalist); and the level of confidence in the journalist personally.


  1. Likewise, when sources do “dry up”, it will often not be clear why. Was it the threat of a subpoena? Dissatisfaction with the quality (or effects) of the coverage? A tactical decision? A mixture of reasons?


  1. It seems likely that journalists’ relationships with their sources has the biggest bearing on the provision of important confidential information. (Blasi 1972, McNamara 2009)

  An alternative?

  1. Some important confidential information may continue to be supplied in other ways even if sources are chilled from approaching the media directly. The Protected Disclosures Act protects whistle-blowing within the government (though not to the media; and there is no absolute assurance of protection of identity). Alternatively, information can be slipped anonymously to the media in an envelope or provided in an anonymous telephone tip. This does, however, make it harder for the media to check.


  1. In the final analysis, my view is that sources are most likely to be chilled if there is a pattern of high-profile court-ordered disclosure (and the journalists comply rather than going to jail or paying a fine), and particular sources regard (sensibly or not) this as something that may happen to them. I say “sensibly or not” because it is the perception of the potential sources that matters. It doesn’t matter whether there is in fact no chance that they will ever be exposed through a court proceeding if they believe they might.


  1. We simply do not have the evidence to assess what goes on in the minds of sources. In my view it is reasonable to guess that (a) sources’ views will differ, (b) most of the ordinary business of off-the-record conversations will be substantially unaffected; (c) some sources involved in leaking significant information, especially where a lawsuit or prosecution is likely to follow, will be concerned if they perceive a real possibility of court-ordered disclosure, (d) if they make a calculation, it is likely to based on their perception of the similarities between their situation and that of other sources whose identities have been ordered to be disclosed, and (e) this will be one of several factors that will influence their decision about coming forward. But I am aware of no empirical evidence to support this analysis.

Sources consulted 

Floyd Abrams and Peter Hawkes “Protection of Journalists’ Sources Under Foreign and International Law” (2004) in Media Law Resource Centre Bulletin, MRLC White Paper on the Reporters’ Privilege August 2004, 183

Agents of Discovery: A Report on the Incidence of Subpoenas Served on the News Media in 1991, a survey by the Reporters Committee for Freedom of the Press (1992). 

Laurence B. Alexander, Linda M. Perry, Bill F. Chamberlin, “Bransbury v Hayes Revisited: A Survey of Journalists Who Become Subpoena Targets” (1994) Newspaper Research Jrnl Spring 1994, 83


Laurance B. Alexander “Looking Out for the Watchdogs: A Legislative Proposal Limiting the Newsgathering Privilege to Journalists in the Greatest Need of Protection for Sources and Information” 20 Yale Law & Pol. Rev. 97 (2002)


Amicus Curiae Brief on Behalf of Various Media Entities and in Support of Jonathan Randal’s Appeal of Trial Chamber’s “Decision on Motion to Set Aside Confidential Subpoena to Give Evidence” August 17, 2002


Merris Amos “A Storm Brewing” New Law Journal 9 August 2002


Australian Press Council Protection of Journalists Sources 25 May 2007


Lillian R. BeVier “The Journalist’s Privilege: A Skeptic’s View” (2006) 32 Ohio Nthrn U Law Rev 467


Vince Blasi “The Newsman’s Privilege: An Empirical Study” (1971-1972) 70 Mich L. Rev. 229


Kelly Buchanan “Freedom of Expression and International Criminal Law: An Analysis of the Decision to Create A Testimonial Privilege for Journalists” (2004) 35 VUWLR 609

 Cohen v Cowles Media 501 US 663 (1991) 

Ruth Costigan “Protection of Journalists’ Sources” Public Law 2007. 464


Ian Cram “Terrorism Investigations and the Coerced Disclosure of Journalists’ Materials” (2009) 14(2) Comms. L, 40


Randall D Eliason “The Problems With the Reporter’s Privilege” 57 Am. U. Law. Rev. 1341 (2007-2008)


Randall D. Eliason “Leakers, Bloggers, and Fourth Estate Inmates: The Misguided Pursuit of a Reporter’s Privilege” (2006-2007) 24 Cardozo Arts & Ent L. J. 385


Factum of the Canadian Civil Liberties Association in National Post v The Queen, April 20, 2009-06-21


Anthony L. Fargo “The Year of Leaking Dangerously: Shadowy Sources, Jailed Journalists, and the Uncertain Future of the Federal Journalist’s Privilege” (2005-2006) 14 Wm & Mary Bill Rts J. 1063


Freedom of Expression Litigation Project Briefing Paper on Protection of Journalists’ Sources Article 19, May 1988


Diana Ginsberg “Moral Imperatives Protected or Punished: Journalists Breaking the Law or their word for their sources’ protection” (2008) City University, London, LLM Paper.

 Hood v TVNZ BSA 2007-028 14 August 2007 ( 

David Cay Johnston “Most Editorss Say They’d Publish Articles Based on Leaks” New York Times July 11 2005

 Diocese of Dunedin et al v TV3 BSA 1999-125 to 1999-137 ( September 1999 ( 

Kathryn M. Kase “When a Promise Is Not a Promise: The Legal Consequences for Journalists Who Break Promises of Confidentiality” 12 Hastings Comm. & Ent. LJ 565 (1989-1990)


Jane E. Kirtley “Will the Demise of the Reporter’s Privilege Mean the End of Investigative Reporting, and Should Judges Care If if Does?” 32 Ohio Nthrn U Law Rev 519 (2006)


Robert D. McFadden “Newspaper Withholding Two Articles After Jailing” New York Times, July 9, 2005


L McNamara “Closure, Caution and the Question of Chilling: How Have Australian Counter-terrorism Laws Affected the Media?” (2009) 14 Media & Arts Law Rev, 1


Susan Nash “Freedom of Expression, Disclosure of Journalists’ Sources and the European Court of Human Rights” (1997) 1 (5) Int Journ Evid. & Proof, 410


Jeffrey S. Nestler “The Underprivileged Profession: The Case for Supreme Court Recognition of the Journalist’s Privilege” 154 U Penn L Rev 201 (2005-2006)

 New Zealand Law Commission, Evidence Law: Privilege Preliminary Paper 23 (May 1994), Ch 13 

Notice of Motion and Motion for Leave to File Brief Amicus Cruiae; Memorandum of Points and Authorities in Support Thereof, In Re Grand Jury Subpoenas to Lance Williams and Mark Fainaru-Wada 2 June 2006


Stephanie Palmer “Protecting Journalists’ Sources: Section 10, Contempt of Court Act 1981” (1992) Public Law, 61


Timothy Pinto “How Sacred is the Rule Against the Disclosure of Journalists’ Sources?” (2003) 14(7) Ent. Law Rev. 170


Geoffrey Robertson, QC and Andrew Nicol, QC, Media Law (Sweet & Maxwell, London, 5ed, 2007)


Geoffrey R. Stone “Why We Need A Federal Reporter’s Privilege” (2005-2006) 34 Hofstra L. Rev. 39


Jeffrey Toobin “Name That Source” The New Yorker January 16, 2006


James Thomas Tucker “Enacting A Reasonable Federal Shield Law: A Reply to Professors Clymer and Eliason” (2007-2008) 57 Am U. L. Rev. 1291


Eileen M. Wirth “Impact of State Shield Laws on Investigative Reporting” 16 Newspaper Research J. 64 (1995)

 Steven D. Zansberg “The Empirical Case: Proving the Need for the Privilege” (2004) in Media Law Resource Centre Bulletin, MRLC White Paper on the Reporters’ Privilege August 2004, 145