Note to trial courts: When ineffective assistance of counsel claims are based what trial counsel said to his client, hold an evidentiary hearing. Note to defense counsel: Data showing the sentences received by defendants charged with the same crimes as your client is about as useful as data showing a patient diagnosed with a lethal illness the survival rates of similarly-diagnosed patients. Note to all: A single, inaccurate, hyperbolic remark during the course of a long sentencing explanation is harmless even if the trial court relied upon it.
Carlson pled guilty to multiple sexual assaults of a child under 16 and received a 23-year bifurcated sentence. Carlson sought to withdraw his guilty plea claiming ineffective assistance of counsel. His trial lawyer told him that pleading guilty would give him the “realistic possibility” of a nonprison sentence. He also claimed that his sentence was based on inaccurate information about the number of times he assaulted the victim, his lawyer was ineffective for failing to object to the inaccuracy, and his sentence was unduly harsh.
Plea withdrawal based on erroneous legal advice
To support his claim for plea withdrawal, Carlson alleged that his trial counsel did not know until sentencing that the total number of assaults might be as high as 300. Had counsel known, he would not have advised Carlson to plead guilty. The court of appeals held this advice objectively reasonable under the first prong of Strickland‘s test for IAC:
As a first-time offender, Carlson was eligible for a nonprison sentence, but the nature of the allegations and Carlson’s confessions foreclosed most favorable sentencing arguments. Among the strongest and best “ammunition” left to Carlson was a guilty plea, which could eliminate the additional trauma of trial and demonstrate his remorse. His counsel’s alleged failure to recognize that the number of assaults would be higher than heretofore recognized was immaterial in this context. A guilty plea was an objectively reasonable strategy for obtaining a nonprison sentence. Besides, how was Carlson’s counsel supposed to know that the court would sua sponte add up the number of assaults at the hearing so as to prepare a response? Slip op. ¶31.
Carslon also argued that “objective sentencing data” showed that a nonprison sentence for the crimes he was charged with was not a “realistic possibility.”
The court will not consider that data at sentencing. So its relevance is limited. The reality is that whatever the objective sentencing data shows about overall sentencing, a defendant’s chances of a particular sentence are individual. A defendant’s situation in trying to predict the sentence he or she will receive is similar to the situation of a person diagnosed with a serious illness. Data may show that only 25% of people survive the disease, but for any particular individual, the chances of survival are black or white, zero or 100%. Likewise, few individuals convicted of both charges Carlson was facing may obtain nonprison sentences, but for any particular individual, the chances were zero or 100%. And in Carlson’s individual situation, if a nonprison sentence was his hope, pleading guilty was a sound strategy—perhaps the only realistic strategy—for achieving that outcome. Slip op. ¶33.
The court of appeals also urged trial courts to hold evidentiary hearings when a defendant claims ineffective assistance of trial counsel based on something his lawyer told him.
We can only picture the angst of dedicated criminal defense lawyers when words they did not say are ascribed to them by default. More importantly, there may come a case where the failure to hold a hearing compromises the trial court’s decision on the matter. By making a record, our court can operate under full knowledge of the facts rather than a supposition of what was claimed to be said. In this case, we had to assume that counsel actually used the term “realistic possibility.” But assuming the truth of that statement, it did not amount to ineffective assistance for the reasons stated. Slip op. ¶35.
Just playing devil’s advocate here. Wouldn’t a “dedicated” criminal defense lawyer instead swallow his pride and admit that he oversold pleading guilty?
Sentence based on inaccurate information and IAC for failure to object
It is well established that a defendant has a due process right to be sentenced based on accurate information. If he claims a violation of this right, then he must prove that the information was inaccurate and that the trial court relied upon it. State v. Travis, 2013 WI 38, ¶17, 347 Wis. 2d 142, 832 N.W.2d 491; State v. Tiepelman, 2006 WI 66, ¶26, 291 Wis. 2d 179, 717 N.W.2d 1. Here, the State conceded that the court relied on its sua sponte calculation that the defendant assaulted the victim “more than 300 times” over the years. The court of appeals held that based on the complaint’s allegations, the math was easy to do, the result was essentially accurate, and any inaccuracy was “harmless.”
¶38 . . . More importantly, even if the “300 to 400” number the court mentioned was to some degree inflated, there is no reasonable probability that such inaccuracy contributed to Carlson’s sentence. It was a single, hyperbolic remark during the course of a long sentencing explanation. As we have already explained, the factors that formed the basis for the court’s sentence were Carlson’s exploitation of his position of trust, his victim-blaming, and the need to deter others from such offenses, not the number of assaults.
¶39 For the same reason, it was no error for Carlson’s trial counsel not to object to the “more than 300 assaults” remark. The challenge for Carlson’s trial counsel at sentencing was that Carlson’s repugnant statements and attitude during the presentence investigation, along with his minimizing of his crimes and blaming the victim, greatly undermined the favorable impact of his guilty plea. In this context it was an objectively reasonable decision to forego quibbling about whether the assaults numbered in the single, double, or triple digits, just before the court pronounced sentence.
Carlson admitted only to 5 or 6 assaults, and the trial court’s math does not add up. See Carlson’s brief at 19-21. Furthermore, the court of appeals seems to have created an exception to the rule SCOW established in Tieplman/Travis. Namely, “[T]he fact that other information might have justified the sentence, independent of the inaccurate information, is irrelevant when the court has relied on inaccurate information as part of the basis of the sentence.” Travis, ¶47. See also Tiepelman, ¶14. So can a sentence based on inaccurate information ever be “harmless”? Perhaps SCOW will have to sort this one out.