The Verdict the Jury Wasn't Allowed to Reach

The Stakes

On April 8, 2026, a Honolulu jury convicted Dr. Gerhardt Konig of Attempted Manslaughter based upon Extreme Mental or Emotional Disturbance, known in Hawaii law as EMED. Neither attorney had asked for that verdict. The defense had fought to keep EMED off the form. The prosecution had attacked it preemptively in closing. The jury picked it anyway.

Sentencing is currently set for August 13, 2026. Under Hawaii law, attempted manslaughter based on EMED is the only Class A felony in Hawaii exempt from the mandatory twenty-year indeterminate sentence. The range is probation to twenty years.

On May 15, 2026, defense counsel Tom Otake filed a Motion for New Trial. The motion runs eleven pages of memorandum plus declarations and four exhibits. It rests on three independent legal grounds. Otake gives Judge Wong three different doors to walk through. He doesn't argue any one of them is stronger than the others. He just makes the judge pick.

Here's the architecture. The three grounds are alternative paths into the same downstream consequence. Once any one of them succeeds, a chain of legal mechanisms triggers automatically and shrinks the state's case dramatically. Same chain. Three triggers.

The judge's choice is which door to walk through. The chain on the other side runs the same regardless. Konig either gets sentenced August 13 with up to twenty years on the table, or the case resets and the realistic ceiling drops to a class B felony assault charge with a ten-year max — and he likely makes bond pending retrial.

The Legal Framework Otake Sets Up First

Before any of the three grounds, the motion spends two pages on Hawaii Rule of Evidence 606. That's not throat-clearing. It's strategic foundation-laying because the state's strongest procedural defense will be HRE 606.

What HRE 606 Does

HRE 606(b) prohibits juror testimony concerning the effect of anything upon a juror's mind, emotions, or mental processes during deliberations. The rule exists to protect the integrity of jury deliberations. Without it, every losing party could pick apart the verdict by hauling jurors in to second-guess themselves.

What HRE 606 Does Not Do

HRE 606 does not impose a blanket prohibition against judicial inquiry into objective misconduct, irregularities, or outside influence. The HRE 606 Commentary describes the dividing line as one between subjective mental processes (off limits) and "conditions or occurrences of events calculated improperly to influence the verdict" (fair game).

Why Otake Goes There First

Otake cites State v. Holbron, State v. Furutani, and Kealoha v. Tanaka to show that Hawaii courts have repeatedly allowed inquiry into objective issues. He's preempting the state's argument that all of this should be inadmissible. His framing is precise: he is not asking the court to crack open deliberations or probe subjective mental processes. He is asking the court to look at what jurors said publicly, what the state itself disclosed, and the objective inconsistencies that emerge.

This matters because it explains why these post-verdict statements are legally cognizable at all. Without HRE 606 cleared first, none of the three grounds work.

Ground One: The Jury Misunderstood the Instructions

HRS § 635-56 authorizes a new trial where the verdict reflects "a misunderstanding of the charge of the court on the part of the jury." HRPP Rule 33 authorizes a new trial in the interests of justice. Both apply here.

What the Jury Was Told to Do

Judge Wong instructed the jury on a sequential decision tree:

1. If you find intent to kill as required for attempted murder, consider whether that intent was influenced by reasonable emotional disturbance.
2. If yes, return Attempted EMED Manslaughter.
3. If no influence by emotional disturbance, return Attempted Second Degree Murder.
4. If you find no intent to kill, skip the EMED instruction entirely and jump to the first assault instruction.

That last directive is the gate. Intent to kill is the threshold question. No intent to kill, no EMED, no attempted murder. Drop straight to assault.

What the Foreperson Said Publicly

After the verdict, jury foreperson Makalapua Atkins gave voluntary interviews to two outlets. Her statements appear in the motion as Exhibits A and B.

The intent to kill is improbable to us based on the evidence presented.

She said the jury concluded "there was definitely a scuffle" that "resulted in injury to the head that could be bodily injury and serious."

Ultimately, we just agreed that she was definitely hurt with a bodily injury to the head that could result in serious bodily injury.

What Juror No. 3 Said Publicly

Juror No. 3 gave an interview to KHON2 News, appearing in the motion as Exhibit C.

I don't know if he was trying to kill his wife. He was emotionally unstable, and his conduct had the capacity to kill his wife.

The Sworn Declaration

This is the part of the motion that should make any lawyer reading it sit up. With Judge Wong's permission, defense counsel contacted the foreperson directly to confirm her media statements were not being misunderstood. She signed a declaration under penalty of law on May 12, 2026.

Paragraph 1 of the declaration identifies her as the foreperson. Paragraph 2 confirms she stands by her media statements. Paragraphs 3 and 4 are the load-bearing language:

The jury did not unanimously believe that Dr. Konig had an intent to kill his wife.

I did not believe that Dr. Konig had an intent to kill his wife during the incident on the Pali Puka trail, and I believed that he was not guilty of Attempted Murder.

Why This Is a Ground for a New Trial

Attempted EMED manslaughter requires intent to kill. EMED is a mitigation of murder, not a separate offense. Hawaii Supreme Court precedent on this is settled across three decades — State v. Sawyer (1998), Whiting v. State (1998), and State v. Adviento (2014) all hold that EMED manslaughter necessarily concedes the defendant intentionally or knowingly caused (or attempted to cause) the death of another.

If the jury did not find intent to kill — and the foreperson has sworn under penalty of law that they did not — they were instructed to skip EMED and return an assault verdict. They did not do that. That is a misunderstanding of the charge within the meaning of HRS § 635-56.

Ground Two: The Verdict Rests on a Nonexistent Offense

State v. Stanley, 148 Hawaiʻi 489 (2021). State v. Loa, 83 Hawaiʻi 335 (1996). Hawaii does not recognize the offense of attempted reckless manslaughter.

The Three Buckets

To understand this ground, hold three categories in mind. Hawaii law treats them as distinct.

• Attempted Murder. You tried to kill someone and failed. The trying-with-intent is the crime. Class A felony, twenty-year mandatory indeterminate sentence.
• Attempted EMED Manslaughter. Same conduct, same intent to kill, same failure to cause death — but the intent was formed under extreme mental or emotional disturbance. EMED is a mitigation that reduces what would have been attempted murder down to attempted manslaughter. Same Class A felony, but exempt from the mandatory twenty-year structure. Range: probation to twenty years. That is the conviction Konig is currently sitting on.
• Assault in the First Degree. You did something so dangerous it could kill someone, you consciously disregarded the risk, and someone actually got seriously hurt. No intent to kill required. Class B felony, ten-year maximum.

The first two buckets require intent to kill. The third does not. That is the dividing line Hawaii law turns on for cases like this one.

The Legal Impossibility

Now the doctrine that makes Ground Two work. Hawaii law does not recognize the offense of attempted reckless manslaughter. You cannot try to be reckless.

The reason is structural, not policy. Hawaii's attempt statute, HRS § 705-500, requires that the defendant act intentionally — that they were aiming at causing a specific result and took a substantial step toward causing it. Recklessness, by contrast, is conscious disregard of a substantial risk. A reckless actor is aware of a risk and proceeds anyway, not caring whether the result happens.

Those mental states cancel each other out. You cannot simultaneously intend to cause someone's death and merely disregard the risk that you might cause it. One requires aiming. The other requires indifference. Together they describe nothing — not any real human state of mind. So Hawaii law treats attempted reckless manslaughter as a legal impossibility. It is not a chargeable offense. It cannot be indicted. It cannot be tried. It cannot be convicted on. The Hawaii Supreme Court in State v. Stanley and State v. Loa has been clear on this for three decades.

Why Assault Catches the Conceptual Territory

This raises an obvious question. If reckless conduct that creates a risk of death is criminally dangerous, and Hawaii law doesn't allow attempted reckless manslaughter, what crime covers it?

The answer is assault. Hawaii's assault statutes are graded by both mental state (intentional, knowing, reckless, negligent) and severity of injury (serious, substantial, ordinary bodily injury). They are explicitly designed to catch the conceptual territory of reckless conduct causing harm short of death. The Hawaii Penal Code Supplemental Commentary describes assault in the first degree as combining the most culpable mental states with the most serious bodily injury 'short of homicide.' That phrase is doing real work. Assault is the residual category — what would have been some form of homicide if the victim had died.

In a case like Konig, where the victim survived and the conduct in dispute is whether striking with a rock reflected an intent to kill or reckless dangerous conduct, the assault framework is the legally appropriate destination for the recklessness reading. Not attempted reckless manslaughter, which doesn't exist.

Why It Applies Here

Read the juror statements through that framework. The foreperson described conduct that 'could result in serious bodily injury.' Juror No. 3 described conduct that had 'the capacity to kill.' The foreperson reiterated, on the record, that the jury believed Konig 'wasn't trying to kill her.'

That is not Bucket One. It is not Bucket Two. Both of those require intent to kill, and the jurors openly disclaimed it. The reasoning the jurors described is Bucket Three — assault. They described dangerous conduct creating a risk of serious harm, by someone who was emotionally unstable. That is the conceptual territory of assault in the first or second degree, not of attempted EMED manslaughter.

In Stanley, the Hawaii Supreme Court reversed a conviction where the record demonstrated a reasonable possibility that the jury had convicted on the legally nonexistent offense of attempted reckless manslaughter. The court held that where the record reflects jury confusion concerning the distinction between attempted EMED manslaughter and attempted reckless manslaughter, the conviction cannot reliably stand.

That is the heart of Otake's Ground Two. The verdict form said attempted EMED manslaughter. But the jurors' own public statements show they were reasoning in the assault framework — recklessness, dangerous conduct, consequence-based. Convicting on a Bucket Two charge while reasoning in Bucket Three is exactly the kind of mismatch Stanley says cannot reliably support a conviction.

Independent of Ground One

Ground One says the jury misunderstood the instructions and convicted on a charge that required a finding (intent to kill) that they openly say they did not make. Ground Two says even if they understood, the reasoning they actually applied does not match the offense they returned — and the offense that does match their reasoning is one Hawaii law does not recognize. The two grounds operate independently. Either one, by itself, supports vacating the conviction.

Ground Three: The Livestream Texter

Remmer v. United States, 347 U.S. 227 (1954). State v. Furutani, 76 Hawaiʻi 172 (1994). Private communications with deliberating jurors are presumptively prejudicial, with the burden of proving harmlessness shifting heavily to the government.

What Was Disclosed

Two days after the verdict, on April 10, 2026, a member of the public emailed the prosecutor's office. The sender had been watching the trial via Law & Crime's YouTube livestream and had noticed a comment in the live chat. The state preserved the email and the screenshot and disclosed both to the defense. They appear in the motion as Exhibit D.

The Comment

During deliberations, while the verdict clock on Law & Crime's broadcast read 3:00:00, a viewer posted in the live chat:

I've been texting with one of the juror's and they're going for count 3 attempted manslaughter.

Three Indicia of Reliability

Otake builds three converging signals that this was not random internet noise.

• The prediction was accurate. The commenter named the eventual verdict before it was publicly announced. Otake cites State v. Prendergast for the proposition that accurate prediction is itself an indicator of reliability.
• The terminology matches. The commenter referred to "count 3." Atkins later referred to the verdict-form options as separate "counts" twice during her East Idaho News interview. Neither the court's instructions nor the verdict form itself used the word "count" for these options. So where was the foreperson getting that vocabulary? It was being used inside the jury room.
• The position matches. The commenter said "Count 3." Attempted EMED Manslaughter was the third option on the verdict form. But the verdict form had not yet been made public at the time of the post. Whoever the commenter was texting with had inside knowledge of how the form was sequentially structured.

Otake's Careful Framing

Footnote 6 of the motion is precise: the motion does not allege that the foreperson was the texter. Any of the twelve jurors could have been the one texting. The foreperson's use of the word "count" simply demonstrates that the term was being used during deliberations among the jurors. That's enough to corroborate the commenter's claim of inside contact.

Why This Is the Headline-Grabbing Ground

This goes beyond Konig. We are now in a world where high-profile trials are livestreamed on YouTube, and jurors are not sequestered. Konig's jury went home every night. If somebody inside that jury room was texting with somebody on the outside who was watching the trial unfold on a livestream — and the state thought the evidence was credible enough to disclose — that is a structural problem for every livestreamed trial in America. It is also, under Remmer, presumptively prejudicial, with the burden shifting hard to the government to prove no harm.

The Chain: What Happens When Any Ground Succeeds

State v. Sylva (2023). State v. Eli (2012). State v. Stanley (2021). Whiting v. State (1998). McElrath v. Georgia (2024).

First, the Important Caveat

Double jeopardy and the chain analysis below only matter if Judge Wong grants the motion. If he denies it, the conviction stands. Konig is sentenced on August 13 with the full range still in play — probation to twenty years. He stays in custody. He pursues a direct appeal through the Hawaii Intermediate Court of Appeals, but the conviction remains in effect during the appeal. The chain never triggers.

The rest of this section assumes the motion succeeds on one of the three grounds. That's the moment the chain starts running.

Editorial Read: Which Door Is Judge Wong Most Likely to Walk Through?

This is editorial. Otake doesn't argue any ground is stronger than the others. He puts three on the table and lets the judge pick. But from the watchdog seat, one of these doors is structurally easier for a judge to walk through than the other two.

Grounds One and Two require the judge to look at what jurors said publicly and apply legal doctrine to their reasoning. That's a precedent-setting move. Hawaii judges generally don't want to disturb compromise verdicts, and they really don't want to set precedent that invites courts to second-guess what jurors meant when they spoke to reporters. The HRE 606 analysis Otake does to clear that path is careful, but it asks the court to engage with deliberative reasoning, even if framed as objective irregularity.

Ground Three is different. The texter allegation is structural, not interpretive. A juror communicating with someone outside the deliberation room — and Remmer v. United States makes that presumptively prejudicial — is the kind of breach a judge essentially has to act on. Granting relief on Ground Three doesn't require the court to second-guess anything jurors said publicly. It just applies a sixty-year-old federal doctrine to an objective set of facts: a comment in the Law and Crime livestream chat, an accurate prediction of the verdict before it was public, terminology that matches what the foreperson later said, and a position match to a verdict form not yet released.

Editorial read: Ground Three is the door Judge Wong is most likely to walk through. It's the path that requires the least precedent-setting from the bench. The other two grounds ask the court to do harder doctrinal work.

But — and this is the architectural point — it does not matter which door the judge picks. The chain runs the same.

The Chain, Step by Step

Once Judge Wong grants the motion on any ground, four legal mechanisms trigger in sequence. None of them depend on which ground was the trigger. The chain is automatic.

Step One: The Conviction Is Vacated

Granting the motion means the EMED manslaughter conviction goes away. The case resets to a pre-trial posture. The state has to decide whether and how to refile charges.

Step Two: The Implied Acquittal Survives

This is where the doctrine starts doing real work. When the original jury returned attempted EMED manslaughter, they had the option to convict on attempted murder and they declined to. Under Hawaii law that declination is an implied acquittal of the greater charge. State v. Eli (2012) holds it directly: 'Because Defendant was found guilty of attempted manslaughter, he was impliedly acquitted of attempted murder in the second degree, and double jeopardy bars retrial for that offense.' State v. Sylva (2023) grounds the rule in EMED specifically.

The state will argue that if the conviction was invalid because the jury misunderstood, then the acquittal it implied should also be invalid. The U.S. Supreme Court already answered that argument in McElrath v. Georgia (2024). An acquittal stands regardless of the procedural posture surrounding it. Vacating the conviction does not undo the implied acquittal on the greater charge.

Result: attempted murder is permanently foreclosed. The state cannot refile it. Ever.

Step Three: Whiting Forecloses EMED Manslaughter

Now the second mechanism. Whiting v. State (1998) holds that EMED is a defense, not a chargeable offense. The state cannot indict someone for attempted EMED manslaughter from scratch. The only procedural vehicle to reach an EMED conviction is to charge attempted murder and have the defense raise EMED as a mitigation. The defense wins on EMED, the conviction drops down to attempted manslaughter.

But the state cannot charge attempted murder. Step Two closed that door. Without the attempted murder vehicle, there is no procedural path back to attempted EMED manslaughter. The law does not allow EMED manslaughter to be charged as a standalone offense. Not because of a policy choice. Because Whiting says so.

Result: attempted EMED manslaughter is also foreclosed. Not 'probably foreclosed.' Foreclosed.

Step Four: The State Is Stuck at Assault One

With attempted murder and attempted EMED manslaughter both off the table, the realistic ceiling for a retrial is attempted assault in the first degree. That's a class B felony under HRS § 707-710. The maximum sentence is ten years under HRS § 706-660, indeterminate, with the Hawaii Paroling Authority setting the minimum.

Below that on the verdict form, the state could also pursue assault in the second degree (class C felony, five-year max) or assault in the third degree (misdemeanor). But the ceiling is class B. Ten years.

The Bail Consequence

This is the part of the chain that has the most immediate human consequence and that gets the least attention in the legal commentary. Pay attention.

Konig has been in custody at OCCC since March 2025. Bail was denied because the original charge — attempted murder in the second degree — was a class A felony carrying mandatory twenty-year exposure. The judge denied bail on flight risk and danger-to-victim grounds, both of which were heavily influenced by the seriousness of the charged offense. Otake tried for bail of $100,000 to $200,000 and lost.

Now imagine the bail picture if the motion is granted. The attempted murder charge is gone. The class A felony exposure is gone. The realistic refileable charge is a class B felony with a ten-year max — and ten years indeterminate, not mandatory. The flight risk analysis changes fundamentally. There is no life sentence to flee. There is no twenty-year sentence to flee. There's a class B assault charge with conditions of release available: no contact with Arielle, no contact with the children, surrender passport, GPS monitoring, residence restrictions.

If the motion is granted, Konig likely makes bond pending retrial. That's the practical consequence of the chain. He walks out of OCCC while the state tries to put together a class B felony case it had not prepared for the first time around.

That is the news the audience can grasp. The legal chain is intricate. The human consequence is immediate.

Why the Architecture Matters

Tom Otake didn't just file a motion with three grounds. He filed a motion engineered so the judge has three different exits and every exit lands in the same place. The judge can grant on the most legally cautious ground available — the juror tampering on Ground Three — and never engage Grounds One or Two. The chain still runs. The state still loses access to attempted murder. The state still loses access to EMED manslaughter. Konig still ends up facing a class B felony at most, with a real shot at bond pending retrial.

That's the lawyering. Three doors. One chain. Same destination.

The Counterargument to Watch For

The state will argue that if the verdict is invalid because the jury misunderstood the instructions or convicted on a nonexistent offense, then the implied acquittal it produced should also be invalid. The McElrath decision is the answer to that argument — an acquittal stands regardless of what happens to the conviction sitting next to it. But the state will push it, and Judge Wong will have to address it.

Read the Motion

Everything in this analysis is built off the actual Motion for New Trial filed by Thomas Otake on May 15, 2026. Read it for yourself. The motion runs eleven pages of memorandum with sworn declarations, four exhibits, and citations to every Hawaii case discussed here. Nothing in this analysis substitutes for reading the original filing.

Download the filing: State of Hawaii v. Gerhardt Konig — Motion for New Trial (Dkt. 419)

This is one watchdog's read of a complex filing. The motion itself is the record.

— END BRIEF —