A gracious reader brought my attention to an article entitled The Silence Penalty, 103 Iowa L. Rev. ____ (forthcoming 2017), which is soon to be published by the University of Iowa College of Law, and which is authored by Professor Jeffrey Bellin of William & Mary Law School. Professor Bellin has examined data from actual criminal trials, as well as the results of a recent 400-person mock juror simulation, to conclude that an accused person who declines to take the stand in a jury trial suffers a conviction rate penalty that is about equal to having evidence of a prior conviction presented against them. In contrast, “for defendants without prior convictions, testifying coincided with an almost doubling of the chances of acquittal.” The Silence Penalty at 26.   Professor Bellin warns:

The surprising power of the silence penalty should give pause to the many defendants without a prior record who demand a trial but then decline to take the witness stand[.] . . . Declining to testify, [] puts them in the same position as a defendant with prior convictions. This is a major blow to acquittal prospects and one that (tactically speaking) should be avoided if at all possible.

Id. at 30. For defendants with prior convictions, the research found that the rate of conviction was about the same whether the accused testified or not.

Though it supports his conclusions, Professor Bellin’s article does not heavily rely on his 400-person mock juror simulation. Instead, the article spends much of its time digesting and analyzing the real trial data collected by the National Center for State Courts in 2000 and 2001. Id. at 20-37. This is fortunate, because a few features of that 400-person mock juror study, which Professor Bellin fully disclosed in his article, arguably weaken that study’s value.

First, the mock jury study relied upon Amazon’s Mechanical Turk employment website to obtain study participants. The Silence Penalty at 17. As Professor Bellin acknowledged, “Mechanical Turk respondents . . . can skew younger, more female and more educated than the population at large.” Id.  Plus, the study’s participants were paid – and paid at a rate that was “high by Mechanical Turk standards, and resulted in an almost immediate acceptance of the survey by the maximum number of respondents[.]” Id. at 17, fn. 96.

Those factors alone could cause someone to question the reliability of the data.  Beyond that, though, the study is also weakened by the fact that the mock jurors participated separately and independently from each other – there was no opportunity for them to deliberate prior to casting their ballots for a verdict.  Further, while each mock juror was instructed that the defendant was not required to testify, they were not specifically instructed to disregard the defendant’s decision not to do so. Id. at 20, fn. 102.

Still, despite these concerns, the article – especially its analysis of real case data – offers a compelling reason for defense counsel to prepare and advise their military clients to testify in their own defense. A 1993 article from Stanford Law School’s Professor Emerita Barbara A. Babcock, entitled Introduction: Taking the Stand, 35 Wm. & Mary L. Rev. 1, seems to agree:

Let me summarize the discussion so far of taking the stand from the defense view. The decision is fraught—but always in the background is the fact that [the defendant’s] failure to testify will increase the already great probability that he will go to prison.

Id. at 15. Professor Babcock continues:

To make this critical and complex decision, the defendant needs help. And if he has a competent lawyer, his possible testimony has been a subject of discussion from very early in the representation. Those who are well-defended rehearse their testimony; the better defended they are the more they rehearse.


If, even after rehearsal, the defendant’s “testimony is weak in substance and halting in style[,]” id., Professor Babcock recommends the defendant still take the stand and face cross-examination. She asserts that, then, the burden falls on defense counsel to rehabilitate her client during argument. Specifically, she recommends urging the jury to consider:

You saw my client, with her sixth grade education, who has never before spoken to an audience in public, on trial in a federal court; and you saw the government prosecutor with her decades of learning, and years of experience, closely cross-examine her.  Yet, in this unequal contest, you never once heard my client deviate from her basic testimony in this case: she is not guilty.

Id.  In Professor Babcock’s estimation, however weak or halting the defendant’s testimony was, in the hands of a skilled advocate, “fleshed out, this argument alone can make it worth the defendant’s taking the stand[.]” Id.


Scholarship Saturday columns over the past 3 months:

• April 8 – The silence of the accused at trial
• April 1 – Forensic science standards beginning to take form
• March 25 – Sex offender registration statutes are not faring well when subjected to rational basis scrutiny
• March 11 – Military Rule of Evidence 513 – “What men value in this world is not rights but privileges.” – H.L. Mencken (1956)
• March 4 – The call for military justice litigation career tracks
• February 25 – Prosecutorial misconduct vs. unlawful command influence—plus another call from academia to strip commanders of prosecutorial discretion
• February 18 – Punishing the cyber thief
• February 11 – CAAF as sentinel species
• February 4 – The ongoing discussion regarding the placement of military prosecutorial discretion

13 Responses to “Scholarship Saturday: The silence of the accused at trial”

  1. Bill Cassara says:

    Very interesting, and confirms what I have long suspected. In a bench trial, the need for my client to testify is not that great. In a panel case, no matter how many times the panel is instructed not to take it into consideration, it is hard to win a “he said, she said” case, when “he” isn’t saying anything.

  2. ry says:

    Not persuasive.  There are too many factors at play that cannot be isolated to enable any meaningful statistics.  For example, how many had confessions, DNA evidence, forensic inculpatory evidence, video evidence, etc.?  Many of the convictions are likely attributable to the weight of other evidence that drove the decision not to testify, rather than because of the singular decision not to testify.  Even if the cases were all equal in type and quality of evidence and all clients equally compelling, the offense at issue can be significant as can the defense. There is a far greater need for testimony to assert self defense when the force used is questionable than in a theft case where there are multiple alibi witnesses or video alibi evidence. This study essentially presumes the cases are equal and the outcome is directly a reflection of one singular type of evidence.  There is so much more at play including the instructions used and how the attorneys present the case, including how they deal with silence during the arguments and use instructions to drive home points, and also how the juries follow their instructions. I could go on and on….

  3. Advocaat says:

    This was an interesting read, Isaac.  The trial defense division from each branch could probably gather data from the past 2 years without breaking a sweat to get a better feel for which types of cases fare better when the accused takes the stand.  I were a client, I’d want to know about this data.

  4. Dew_Process says:

    There is a fair amount of social science research on this topic, which should be pursued as Advocaat suggests and incorporated into basic defense counsel training. But, this is a good and timely look at the topic.  And, it’s not an unknown subject at SCOTUS.
    In United States v. Grayson, 438 U.S. 41, 58, n.5 (1978), Justice Stewart noted in dissent:

    “A defendant who does not take the stand will probably fatally prejudice his chances of acquittal.”) (quoting Note, The Influence of the Defendant’s Plea on Judicial Determination of Sentence, 66 YALE L.J. 204, 212 n.36 (1956)).

    See also, Lakeside v. Oregon, 435 U.S. 333, 340 n.10 (1978) (recognizing that the negative inference a jury may draw from the accused’s decision not to take the stand may be inevitable.”  And also, Babcock, Introduction: Taking the Stand, 35 Wm. & Mary L. Rev. 1 (1993), available HERE.
    And I think anyone who has had the opportunity to interview jurors / members after a guilty verdict where the Accused did not testify, will generally agree that they tend to ask, “why didn’t the defendant testify?”

  5. stewie says:

    I’ve always thought that panels give most enlisted the benefit of the doubt,  but not so much to officers.

  6. Aquemini says:

    Disclaimer:  I’m not a lawyer.
    Long time lurker, first time poster.  I was the accused in a GCM earlier this year.  I was found not guilty of an Art 120 charge, but guilty of a lesser charges (Art 93/Art 92(3).  During my trial I was advised to not testify because it was a literal “she said/she said.”  If I was to testify I would just confirm some facts for the Government (were you in the room, where you having sex, etc).  I never made a statement to any investigators, never made a statement to my CO, and the there was no physical proof of the alleged assault (never reported it).
    During Voir Dire the question of “would you hold it against the accused if he did not testify” was asked.  All the potential panel members of course said no, however there were a few “well I wouldn’t, but I mean I would want to testify if I was the accused and tell my side of the story.”  The panel was made of all Officers, I’m Enlisted.  The feedback both sides got at the end of the trial from the panel was not testifying on my behalf was extremely smart.

  7. Bionic Barry Dylan says:

    While I will agree this is an interesting read, I think it’s hard to reliably measure this sort of thing.  As a DC, my starting point before really getting in to the meat of a case, was that I wanted my client to testify if possible (especially in a 120 case) to get his/her side of the story out there because, as already mentioned, people are people.  No matter how many times the panel is told not to hold silence against the Accused, they will always be wondering.
    However, there were plenty of times when there was information that only I had because of my exclusive access to the Accused (assuming he/she was telling me the whole story) and the government’s failure to do its job, and plenty of those instances ended in acquittals or other favorable dispositions without my client ever saying a word.
    I think most people would agree they want to hear the Accused’s story, but generally there are only 2 people who have all the facts: the Accused and his/her Counsel.  That makes this kind of thing hard to generalize.

  8. stewie says:

    I don’t think the purpose here is to generalize in an always or never do this or that type way. It’s just a reminder that the common DC tactic of “never have your client testify unless you have no choice/are going to lose for sure otherwise” may not always be the smartest approach in every case.
    The corollary is that you should always PREPARE you client as if they are going to testify so you are ready, just in case stuff is going south and now you aren’t throwing your client up there last minute with no preparation in a desperate attempt to stave off a conviction.

  9. k fischer says:

    I agree.  The DC starting point should be to put your client on the stand.  It is only in rare circumstances that a client whom the DC believes is innocent should not be put up to tell his side.  I’ve never been concerned about my client being cross examined when I believed he was telling the truth and what he was saying showed that he was innocent.  I have many times seen TC’s scrambling for their notebook when the MJ instructs the defense to call it’s first witness and I respond, “The Defense calls [the accused].”  I’ve seen crosses that do no real damage to the Defense’s case, and that is a really powerful display to a panel when your enlisted client is able to go toe to toe and answer the TC/SVP’s questions honestly and respectfully, while remaining relatively unscathed in the joust.
    Of course, I get to pick and choose who I represent, so I pick those who have a valid defense that I can argue, whereas your average detailed counsel does not have that luxury.  So, I have found that the fact that me and my client are the only ones who know his story is an asset.  If it is a valid story, then obviously the TC is going to have a more difficult time preparing on the fly if there is something that he was not aware of that he could not have known, failed to fully investigate evidence that he should have known, or fully investigated the evidence, but failed to analyze it objectively.
    And, just a side question: has anyone felt a “Trump” effect in any panel cases?  That could be a panel who compensates for the fact that their CIC was characterized as a sexual predator by convicting when the facts clearly supported a finding of not guilty.  Or, it also could be an all male panel who seemed to feel relief that they no longer had to feel guilty about being male after the “I’m with Her” candidate was defeated and acquitted. 
    I may have felt the latter in my last case with an all male panel.  I actually had a panel member in an all male panel raise his hand in general voire dire when I asked if “any panel member thought that there was an epidemic of false allegations of sexual assault in the military.”  He went on to say on individual voire dire that he commanded a training unit and was surprised how many of the gender integrated trainees were making false allegations to get out of the Army or switch companies.  (For some reason the Government did not kick him.)  Over the past 8 years, I have thought having an all male panel is bad when you don’t have a female who can explain to them that its okay to be critical of the complaining witness’s story.  This panel seemed to be liberated.  And, my client testified and was acquitted, although I thought the TC (for some reason not the SVP with far more experience) who crossed him did a really good job with what he had.  Then again, it could just be that the last two panels have been really good about following the judge’s instructions and understanding reasonable doubt.

  10. k fischer says:


    The corollary is that you should always PREPARE you client as if they are going to testify so you are ready, just in case stuff is going south and now you aren’t throwing your client up there last minute with no preparation in a desperate attempt to stave off a conviction.

    I agree with this statement that you should at least prepare your client to take the stand by questioning him prior to trial in a mock trial setting, so he can get a feel of what it is going to be like testifying on direct and cross examination.  I can’t remember which Pozner/Dodd seminar I saw, but the decision to put your client on the stand should be made prior to voire dire.  If you are putting him on the stand as some sort of “Hail Mary” pass because the case is not going well, then you’ve most likely lost.
    The panel’s view of the case and the defense counsel’s closing argument differs depending on whether or not the accused takes the stand.  If the accused does not take the stand, then the case is about whether the Government has done its job and met it’s burden.  If the Accused takes the stand, then the case typically is more about whether the panel believes the accused, rather than whether the Government has met it’s burden.   

  11. AF Capt says:

    I never put a client on the stand and have avoided all 120 convictions (except a single plea).  Often times these facts provide a good rationalization that can be planted in voir dire about why a person wouldn’t want to testify.  (May involve copping to underage alcohol consumption, admitting adultery, etc.)  I think members get it when you bring it up that way, early on.

  12. stewie says:

    Article 88 forces me to ignore much of what you wrote in your first response kf except to say I hope panel members are not thinking about who is or is not President when deciding pretty important decisions like guilt or innocence at trial.
    As for Hail Mary’s…it may not be a “Hail Mary” to change your tactics and put your client on the stand at trial. This is particularly going to be true as alleged victims give less and less information prior to trial. Some AVs may be better than others at presenting their story well or addressing things you think are errors or lies or inconsistencies. You won’t know the skill of that until cross is done.
    If they do poorly, then no need to put your client on the stand, if they do well, then you may need to put your client on the stand.
    I also agree that few TC are prepared to actually cross an accused. They rarely expect an accused to testify so they don’t spend time preparing for it, and an unprepared counsel is usually an ineffective counsel except for a few rare shining stars.
    I think the point is to not be wedded to I never do this or I always do that or I make my decision at this point in a trial no matter what.  You should always be willing to reassess in real time so long as you don’t overreact to things.
    Unless your guy is simply going to lie, or is going to have significant credibility issues, or is going to provide the missing link the prosecution needs, you should be prepared at least for the idea of having them testify…and quite frankly prepping them for direct and cross can do nothing but better help you understand the case even if they don’t testify.

  13. Jason Grover says:

    I’d agree with Bill’s comment that bench trials may be very different with a military judge that knows the rules, may have represented defendants, and likely is much less reflexive in holding it against an accused.

    I would often tell my clients that there was just something in looking the panel in the eyes and telling them your side. I think it is even truer in boards of inquiries. I’m not sure I ever lost a BOI representing the government when the respondent didn’t testify. There was just a big sense of “you’ve got to tell your side of the story.”

    But what made me want to post is I completely agree with the sense that you could catch a TC flat-footed. Cross-examination is a skill. And cross-examination of the accused is its own sub-specialty. And like any skill, it improves with practice. Only there are not that many opportunities to practice it in a world of declining caseloads. As a young TC, I did my best to over-prepare, like any fresh, eager attorney. As I got experience, I had to fight that thought that preparing for the Accused’s cross-examination was a waste of time, because “there’s no way they’ll put him up.” I was caught flat-footed a time or two and those crosses sucked. I think, in general, defense counsel get better cross-examination practice because they are always crossing victims and witnesses. And I’ve seen a fair amount of TC’s overdo the cross-examination of the accused. Going for blood and treating the accused like the devil only works when they are the devil, and the overwhelming majority of military accused are not the devil. 


    So I’d tend to agree, in most cases, I think the defense is in a better position when they put the accused on the stand and, in general, the government is prepared for what it wants to happen and what it expects to happen. That is why being a defense counsel can be so fun, you can alter the script.