This podcast episode accompanies the article from Ms. Vendzules: Guilty After Proven Innocent — Hidden Factfinding in Immigration Decision-Making.

Transcript

SPEAKERS

Peter Mason, Sarah Vendzules

Peter Mason  00:00

Immigration adjudications regularly use information from the criminal legal system to justify a discretionary denial of relief or benefits, even when charges have been dismissed. This practice faces little scrutiny due to the assumption that adjudicators are merely importing facts already found by the criminal system. But what if this practice actually constitutes hidden fact finding? What framework could bring it back within the bounds of the law. My name is Peter Mason, and you're listening to the California Law Review podcast. Our goal is to provide an accessible and thought provoking overview of the scholarship we publish. In today's episode, we will be discussing the article Guilty After Proven Innocent: Hidden Factfinding in Immigration Decisionmaking, a piece by Sarah Vendzules, Senior Staff Attorney at the Office of the Appellate Defender in New York City. This piece was published in Issue Three of Volume 112 in June 2024. Sarah Vendzules, thank you so much for taking the time to talk with me today about your article. 

Sarah Vendzules  00:55

Thank you, Peter. Thank you for taking the time to interview me. I'm so honored to be here.

Peter Mason  00:58

To begin with, can you summarize your main arguments in this piece?

Sarah Vendzules  01:03

So basically, my argument is that when an immigration decision maker is dealing with an allegation from the criminal legal system and using that against an immigrant, that adjudicator is necessarily finding that the allegations were sufficiently reliable to justify the negative decision if the allegations haven't already been determined to be true by the criminal legal system. But immigration adjudicators frequently don't act like they're engaged in fact, finding instead they sort of tend to accept all the allegations at face value, and say this is a negative factor and then deny relief based on that. But what's obscured by talking about balancing factors in aid of a discretionary decision is a fact finding process, and one that was grossly unfair because the allegations were given the weight of criminal convictions. So you have an immigrant who is not just forced to relitigate their criminal case, but they're forced to do it in a forum where they bear the burden of proof, and also where the decision maker is always going to believe the allegations over any evidence that they can realistically produce. So in the scenario when there's no way that the immigrant can actually refute the allegations, the allegations are essentially given the treatment that you'd expect for criminal convictions. So basically, my argument is that you should treat fact finding like fact finding, and you have to do it fairly.

Peter Mason  02:40

What motivated you to write this article?

Sarah Vendzules  02:42

So I found myself in a unique place as an immigration lawyer, working in a public defender office. Because of my job, I was pretty familiar with the documents that come out of the criminal legal system, including, for example, a criminal complaint or a police report and what they're worth, which is not very much in the criminal legal system. And then practicing in immigration court and before United States Citizenship and Immigration Services or USCIS, the exact same document that is not even reliable enough to be admitted into a criminal case, is treated very differently. There's almost a reverence given to these products of the criminal legal system. Everything that comes out of it has this gloss of reliability. So I spend a lot of time explaining the immigration system to criminal lawyers, and they would usually be incredulous about what I was telling them. I would say things like, well, the immigration adjudicator is going to be looking at these allegations, and they're going to use them to find that this was a particularly serious crime, or they're going to use them as a basis to deny a benefit. And the criminal defense attorneys who know the value of these documents would be saying, Well, that can't be correct. You know, you must be mistaken. So one case in particular that struck with that stuck with me was the client that I mentioned at the beginning of the article. The unfairness of her situation was was really, really stark to me. She'd been through so much. She was a victim of very serious crime. She had cooperated with law enforcement. She had been given this humanitarian visa, the U Visa, and then, you know, when she's now going to convert that into a permanent green card. Allegations from a criminal case are being used against her. And what had happened was, you know, a relative's abusive husband had said something to a police officer that got written down. You know, now it's in a criminal complaint, but there's no testing of the allegations at this stage, you know, the person says something, the officer writes it down, and then if it makes out a crime, the DA charges. The testing that would that the criminal legal system would do, that was the dismissal of the charges. The charges were dismissed. But that work that the criminal legal system did was not something that the immigrantion system gave any value to. The charges had been brought against her, the Immigration System had the complaint, and now she was the one who had to disprove them. And how is she supposed to do that? There didn't seem to be a way that she could do that. So again, my thought was, well, this, this can't be right. So I went and started doing my research, looked into, you know, what is the law here, exactly? I was pretty new as an attorney at that point, and I was really shocked by how anemic the law was in this area. So in the discretionary area, it seemed like immigration can pretty much use whatever information they want against an immigrant, and as long as they call it balancing factors, that's fine. So this article comes out of that, out of my instinct that there was something wrong, and I wrote it in the process of trying to figure out exactly what that was. 

Peter Mason  05:50

So your article begins by laying out some concepts that help set things up. Specifically, you discuss what you call outputs of the criminal legal system, the portability of those outputs, and how they factor into the two steps that most immigration adjudications entail, a non discretionary eligibility determination followed by a discretionary determination about granting the requested relief. Can you walk listeners through those concepts and how they fit into your broader argument?

Sarah Vendzules  06:14

So the term output roughly corresponds to a document from the criminal legal system, like a complaint, a police report. But more specifically, it's really the individual pieces of information that are found within those documents, rather than the documents themselves. So for example, with a criminal complaint, there might be charges like maybe there's three charges on the complaint, and on one of them, the person pled guilty and was convicted of the of the charge, and the others, they did not. So those clearly have very different evidentiary values. And then the idea of output portability is really just, you know, can you use an output in your system if you're an adjudicator, and if so, you know, for what purpose can you use it? So it's not just, is it admissible? Because in the immigration system, almost nothing is going to be excludable. But if you're going to consider it, is it something you can take as true because someone else has already done the work for you, or are you going to have to engage with it yourself? So there's also the concept of discretionary versus non discretionary. Almost every immigration decision has a discretionary component to it, so in the first step, immigration adjudicator is asking if the person is eligible for the relief, eligible for the benefit. And then in the second step, they're asking, do they deserve it, in sort of a exercise of administrative grace. And a lot of attention in the literature has been paid already to step one, but much less to step two, and this is the reason it's pretty obvious that the first step you have a lot of litigation, a lot of fireworks, a lot of Supreme Court cases. In Crim-Imm, for example, advocates are arguing about, you know, is this conviction actually an aggravated felony? Is it a controlled substance offense, which make somebody deportable or not? So there's a lot of law in this area. There is a lot less litigation, a lot less law to be found at step two, because what law are you even arguing about? So that is the basis of this article really is to explore that discretionary gray zone.

Peter Mason  08:15

You get into your argument by laying out how the Board of Immigration Appeals, or the BIA, has provided inconsistent guidance on how immigration adjudicators should use criminal legal system outputs. Can you lay out that background?

Sarah Vendzules  08:27

The first cases where the BIA is saying that it's okay to use criminal legal system outputs in aid of discretionary decision making, the BIA actually acknowledges that the immigration judges and USCIS adjudicators are going to be engaged in fact finding based on these outputs. But that really quickly falls off. So you have two cases within two weeks of each other that are quite good. In one of them, it was a conviction on appeal. So because the conviction was on appeal, the criminal legal system had not completed its fact finding process. So for that reason, the BIA is acknowledging that this output is going to be used in aid of fact finding, but the facts have not already been found. And they talk about evaluating the strength and the nature of the evidence and determining its probative value and its weight. In another case, the IJ had relied on a police report in a case where prosecution had been declined, and the BIA said that you gave this police report too much weight. And they say in that case that these allegations should not be credited unless they are corroborated. So I think reading these two cases together, you can come up with sort of a rudimentary guidance that maybe something corroborative is happening along the way to a conviction. And then, of course, the most corroborative action that the criminal legal system could perform would be actually convicting the person of the crime. But in the steps along the way, the output may in some way, shape, or form gain weight or become more become more probative. So after establishing that guidance back in the 90s, the BIA immediately completely throws it out. So in later cases, you find them discussing the conduct that's described in the police reports as if it's accurate, as if it's an accurate depiction of what has happened. You see phrases like problematic pattern of behavior, which is what they use to describe my clients, allegations. And rather than looking at corroboration or the stage to which the proceedings have progressed, they're looking to other things. So in some cases, they talk about crediting allegations because they're quote, specific and detailed. And I'm not sure if that's a mark of accuracy - it sort of seems a little bit like pop psychology, you know, the idea that, well, this is specific and detailed, it must be true. In other cases, they talk about the immigrant's failure to refute the allegations, as if that were a mark of reliability. And then in other cases, they just say, well, the IJ found the complaint was reliable, and they leave it at that. And they don't engage in any analysis as to what the IJ was doing and why they said it was reliable. So they're not really acknowledging sometimes they go back and forth, but in a lot of cases, they really don't seem to be acknowledging the fact finding that these adjudicators are doing. They talk a lot about balancing factors rather than finding facts. They always just kind of retreat to this idea that the IJ has discretion to balance the factors as they wish. And then, you know, you'll find quotes from the cases that I think are particularly problematic. For example, they'll say things like, Well, the fact that the applicant has not been convicted of the underlying crime does not equate with the finding that the underlying conduct in question did not occur, and USCIS may take into account all factors in making its discretionary determination. So in other words, what they're saying, which isn't made explicit, is that USCIS is making this finding that the conduct did occur.

Peter Mason  11:58

The result of the BIA's guidance and case law, you argue, is burden multiplying from respondents who adjudicators expect to disprove claims that come from unreliable criminal legal system outputs. When respondents face this burden, you call those claims zombie allegations, allegations that, like zombies, rise from the dead and are nearly impossible to kill. Can you explain this burden multiplying effect for listeners, and how it reflects what you call impoverished decision making by immigration adjudicators?

Sarah Vendzules  12:25

When applying for discretionary relief, the immigrant bears the burden of proof that they are eligible for and that they deserve the benefit. This isn't particularly exotic. Other legal systems deal with burdens all the time, including burdens that shift during the course of litigation. But it's not just a shifted or heightened burden, because all the evidence that you as an individual could possibly submit is going to be discounted. The only evidence that the immigration system is actually going to take as credible is evidence from the criminal legal system itself. So for example, the case I discuss where the immigration adjudicator is asking the immigrant to provide a police report, and their stated rationale is that they want to confirm his story. Well, the police report is not going to confirm his story. There's no such thing as a police report that says, Oh, actually, this person didn't do anything wrong. And you know, I think it's important to point out that evidence of factual innocence to the needed standard, which the immigration system is often clear and convincing evidence. That's just not what the criminal legal system is designed to produce. So for that reason, I call it a multiplied burden. I was looking for a word to describe something that's not just a regular burden, not just a shifted burden or a heightened burden, but something else entirely. And then, similarly, with zombie allegations, the term is a little bit dramatic, but I do think it's accurate to the experience. The result for an immigrant is that you're arrested, the allegations have been disposed of, the case was dismissed, or maybe you pled guilty to a lesser charge. But now in the immigration system, the allegations that have already been put to bed are back, and they are in some ways like zombies, because they have arisen from the dead, and they are now effectively unkillable. So one of the examples that I give in the piece is about a trafficking victim who was denied relief because, you know, she had been arrested for prostitution. Again, she's a trafficking victim, and she's denied relief because the immigration adjudicator says that her explanation of what happened differed from the police report, and therefore she's failed to take responsibility, quote unquote, for her actions. Or, you know, the case with my client, where they're talking about her quote unquote problematic pattern of behavior, and I'm asking, what behavior? Where are you getting this from? I mean, obviously that stuck in my craw, but it was something that, you know was to me, quite stark. So you'll have the immigration adjudicator taking the police report, the criminal complaint, whatever it is, the document from the criminal legal system, and assuming that is true, and any evidence that the immigrant submits that contradicts with that is going to not be believed. So the most stark example, I think, that I identified in my piece was the case of the dad who was arrested because his daughter complained that he had hit her. So in the immigration proceeding, he was able to submit a sworn affidavit from her saying I was going through a troubled time, I was a teenager, you know, he didn't want to let me see my boyfriend, something along those lines, and I made it up. But instead of believe what I think is objectively a much stronger piece of evidence, you know, a sworn statement under penalty of perjury. Instead, the adjudicator chose to believe the police report, which, again, is something that is not even reliable enough to be used in court. So I think the major problem with what's going on here, and I think this, this case really lays it out, is that the Father has submitted the strongest possible form of evidence that he could submit to the immigration system to disprove the allegations. It's a sworn statement from the original complaining witness saying it's not true, I made it up. And yet, the immigration adjudicator is giving so much weight to this police report, which is a product of the criminal legal system, that they're discounting his evidence, and they're choosing to believe the criminal legal system output, even though, by any measure, it is objectively much less reliable than the sworn statement under penalty of perjury submitted by the daughter. I really thought that I had found something of a, I guess, a smoking gun with that case, because it just seemed to lay out the unfairness so starkly. And, you know, I might have assumed that maybe cases where this is happening this explicitly are rare. But actually, I was on a call with some advocates just two weeks ago, and they were discussing this exact thing happening in the context of bond hearings where they had a client who had an arrest, and they actually were able to get the complaining witness to make a statement, recanting the allegations. And the immigration judge still denied bond, saying, essentially, well, you wouldn't have been arrested if you didn't do anything. I thought this comment represented a rather stunning level of naivete about the criminal legal system and how it operates. So basically, what this is all accumulating in this assumption of reliability given to criminal legal system outputs, this refusal to acknowledge fact finding as fact finding, but instead to just think of it as a balancing of factors. And then this sort of credibility or trust in anything that comes out of the criminal legal system is that arrests are essentially convictions. I don't think there's any other I don't think there's any way around it. If the allegations are going to be taken as true until proven to be false, and there's no actual way that an immigrant can prove them to be false, then I don't know what else to call it.

Peter Mason  17:36

So these problems with the current practice of immigration adjudication set up your solution: the concept of output portability. Adopting this concept, you argue, would helpfully constrain immigration adjudicators' uses of criminal legal system outputs, while also making their reasoning more explicit and thus more reviewable on appeal. Can you explain your concept of output portability and those immediate implications of it?

Sarah Vendzules  17:57

Sure. So the output portability framework that I lay out in my piece is a way to basically evaluate what outputs could come in and what purpose you could use them for. So it distinguishes between outputs that I call fully portable, which you can just accept with no fact finding, and partially portable, which can be used but have to be used in aid of fact finding in your system. So an output that is partially portable tracks with evidence that would be admissible in a civil case or a criminal case. So you can't treat these facts as if they've already been found; you have to do your own fact finding based on the output. And then finally, the third category would be an output that is nonportable, and this would track with something that's inadmissible. But you know, in immigration proceedings, which are administrative proceedings, we don't really have laws of evidence, and pretty much anything can come in, so it's very rare, or maybe even non existent, that you would have an output that was considered non portable or inadmissible. Almost everything is going to be either fully or partially portable. And that's why I created this framework, is to distinguish between those two things. So to determine whether an output is fully or partially portable, immigration adjudicators, which would be like an immigration judge or a USCIS employee, they would look and see whether the output has been adjudicated, and if so, to what standard. So, for example, in the criminal system, a conviction is considered to have been adjudicated beyond a reasonable doubt, guilt has been established beyond a reasonable doubt. That's the standard, it's a pretty high one. So then in immigration court, immigration only needs to prove that somebody is deportable by clear and convincing evidence. So beyond a reasonable doubt is a higher standard. For that reason, they can have a criminal conviction, and they can take it and they can use it. It's fully portable. There's no need for any further fact finding on the conviction. But if the criminal legal system has found the facts to a lower standard, for example, many restitution orders are found to a preponderance of the evidence, which is lower than clear and convincing, then it's not going to be enough to just accept those facts that were found to a lower standard, give them a promotion and say, oh, actually it is clear and convincing evidence now. You actually need to look at the underlying information that forms the basis of that report, and you need to make your own determination. And then a lot of outputs from the criminal legal system that aren't guilty verdicts or that aren't restitution orders haven't been adjudicated at all, so no facts have been found, there's no standard, so you certainly can't just import those without further finding a fact in your system. And I think this logic sort of underlies why we're able to use criminal convictions in our society in a lot of places. They have been found to this extremely high standard beyond a reasonable doubt, and therefore they can be used in other places without further fact finding. But I do want to clarify here that I do not mean to say that criminal convictions are necessarily, inherently reliable. I know a lot of advocates will take issue with that, for good reason. We have a system that is designed to encourage guilty pleas through extremely stacked incentives. So you know, if you plead guilty, you will get a much, much lighter sentence than you would get if you choose to exercise your right to trial and are found guilty. So you know, in my early training as a criminal defense attorney, I spoke with a number of senior attorneys who, all of them said that without a doubt, if they were accused of a crime, even if they were innocent, they would plead guilty. And that really stuck with me, really struck me. So again, I do not mean to say that criminal convictions are inherently reliable, but I think the distinction that I want to make is that these criminal arrests, documents from the criminal legal system that are not convictions, that do not rise to the level of convictions, are much less reliable even than criminal convictions. And so the point of the output portability framework is to acknowledge the difference between the two, the difference between what has been adjudicated and to what standard and what is just an allegation. And if you're going to be using those allegations in your system, then you need to be explicit about what you're doing, and you need to take ownership of the fact that you are the one who is going to be engaged in this fact finding process. It hasn't already been done for you. You have to do it.

Peter Mason  22:14

You end the article by exploring some broader implications of your output portability framework for upending the current regime of hidden fact finding. The one you spend the most time on is how output portability could facilitate due process claims in the immigration context, most notably in the case of bond proceedings. Can you explain your vision and arguments on that point?

Sarah Vendzules  22:33

Sure. So, to understand due process, you have to understand that there really isn't any other avenue of review of discretionary immigration decisions. So immigration law strips judicial review of almost all discretionary decision making from the courts, and the only exception is for questions of law and constitutional claims, and that's only because they were forced to put that back in. But a question of law is really difficult, because what law are you talking about when you're talking about discretion? Really, the only claim that you have when you're trying to challenge a discretionary decision is a due process claim. But in cases where immigrants bring due process claims, the courts are really struggling to identify the due process error. So you know, they'll talk about the immigrant trying to couch a quibble over fact finding in a constitutional claim. You know, they'll characterize their argument as a mere quarrel over the correctness of the fact finding, and they'll say things like, well, it doesn't categorically violate due process to value one form of evidence over another form of evidence. I think that the due process error is that adjudicators are finding facts, but that this fact finding is not being done fairly. So it's not just the result of the fact finding was wrong, but it was the process. And I think that there are some very basic rules for fact finding, for it to be done consistent with due process. And obviously the sort of bedrock is the opportunity to be heard. So the opportunity to be heard doesn't mean just they're going to listen to you and do what they want anyway. It means that you have to have a meaningful opportunity to affect the outcome of the proceedings. We have to actually make a determination that this allegation is true. It doesn’t necessarily have to be beyond a reasonable doubt or even clear and convincing evidence, but it has to be found to some standard that is sufficient to justify a negative decision against the immigrant. So due process might not require an immigration judge to credit an immigrant's testimony over a given document, like a police report, but it does require the immigration judge to conduct a fair review of the evidence, such that, for example, respondents sworn testimony, if found credible, has a chance to actually overcome the allegations. So for this to happen, you have to treat the allegations in a police report as allegations, not as already adjudicated facts. So in the example, with a father who was arrested after the argument with a teenage daughter, it's really hard for me to see how he was given a meaningful say of the determination of the facts related to his application. The agency is making this finding a fact, but they're not making it to any standard, and it's basically predetermined from the outset because of what I call burden multiplying and zombie allegations, where it's essentially impossible for him to rebut the output. So that's why I think that the output portability framework and the identification of hidden fact finding helps to make it clear that this is not just a quarrel over the weight assigned to any particular piece of evidence, but it's a real issue with the process by which this fact finding is being conducted. So you highlighted bond. Bond is extremely different from most of the adjudications that I talk about or most immigration adjudications. So the consensus is that immigrants don't necessarily have any protected interest in most forms of immigration relief. So there's no liberty interest, there's no property interest. Because of that, when you're doing a due process analysis, you're not balancing the immigrant's interest versus the government's interest. The Immigrant is considered to have no interest in what they're seeking. So the only question is, you know, is this process, quote unquote, fundamentally fair? So that's a pretty low standard. But bond is not like that. Bond squarely implicates the liberty interest, and for that reason, you have a more robust test. You have the Mathews v. Eldridge balancing test, where you compare the government's interest in their process with the immigrant's interest in their freedom, and you look and you see if those are fairly balanced. So the way that the bond system currently operates is very interesting, because immigration detention is civil detention, civil preventative detention, and that's extremely disfavored by our legal system. And the Supreme Court has examined the detention of immigrants and has found that it is permissible to detain somebody (and this is coming out of principles from other types of civil preventative detention) so long as they present a risk of flight or a danger to the community. So in the immigration system, there are certain types of detention that are considered, quote unquote, mandatory detention, and that's for immigrants who have certain convictions. That's been found to be okay, but that's only because these convictions are able to serve as a stand in for a finding of dangerousness. So the court has said that, essentially, we don't have to do individualized determinations of dangerousness, we're going to use these convictions as a stand in. That's what Congress wanted, and we're going to defer to that, and we think it's sufficient. But you know, for people who don't have convictions, again, I'm talking about discretionary adjudications, the requirement still exists that they have to be found to be either a flight risk or a danger to the community. And what's happening currently in bond proceedings, which, again, I'm talking about discretionary detention for people who are in the process of removal proceedings. What's happening in these proceedings is that immigration judges are making findings of dangerousness, and oftentimes, if somebody has not been convicted, they're basing it on allegations from a criminal complaint or from a police report. So I do think that the output portability framework and the identification of hidden fact finding is very relevant here, because if you look at this as a purely discretionary determination, which is how immigration sees bond proceedings, then, because of jurisdiction stripping the courts don't have any opportunity to review these determinations. But you know, if you look at it rather as a factual eligibility determination, is this person actually eligible to be detained because they've been found to be a flight risk or a danger, and then with a second discretionary step of, okay, now that I found this sort of threshold requirement, am I, in my discretion, going to release them, or am I going to, you know, set bond, or even, you know, force them to remain detained without any possibility of paying bond to get out? Then it seems as if that first step, factual eligibility determination by rights really should be subject to judicial review. Disaggregating fact finding from discretion, once you realize that when you're looking at a criminal complaint or looking at a police report and saying, I am finding this is sufficiently likely to be true to justify my decision that this person presents a danger to the community, then that fact finding should get review. So the reason why I think that judicial review of immigration judge determinations are so important is because right now, these determinations, I guess as in the example that I discussed earlier where the immigration judge basically assumed that if the person had been arrested, they must have done something, what's going on is really a very sloppy fact finding process, or, I guess, impoverished decision making process, as I call it in the article. All of the errors that I have mentioned previously, of attributing this presumption of reliability to these allegations, refusing to credit the immigrant's evidence if they contradict the allegations. You know, in some cases, immigration refusing to bring the immigrant to court, to even contest the allegations, to actually have a chance to fight against them. All of these errors are happening in these proceedings, and they're not like other immigration proceedings, because, you know, instead of denying somebody a benefit, which you know they may have no entitlement under the law to receive, what you're doing is you're depriving somebody of their liberty, and you're doing it for months or even years. And so for that to happen, and all of it to be hinging upon this finding of dangerousness that's based on incredibly shoddy fact finding, I think that is it does quite clearly present a potential due process issue. I think you can also make an argument that the burden really should be with the agency from the beginning, with immigration from the beginning. And you know the reason for that is if the Supreme Court is essentially writing a fact finding requirement into the law and saying that in order for detention to pass constitutional muster you have to have this determination of flight risk or danger, then I don't think it's likely that they intended that the immigrant would bear the burden of that. Given the liberty interest involved, given what is at stake, this is not like other types of immigration adjudications. This is about the deprivation of liberty for somebody for months or even years. So I think a lot of these arguments have been made by advocates and are being made, but I think that the contribution of my piece, at least what I hope it is, is forcing adjudicators to be really precise about their fact finding. And I think that does have implications for how we look at bond proceedings as a whole, and how we look at any type of discretionary adjudication, but you know also how we look at bond proceedings, and maybe starts to call into question the assumption that these are purely discretionary determinations.

Peter Mason  31:59

Other implications of your output portability framework include its relation to existing law on the probative value assigned to criminal legal system outputs, as well as its relevance to adjudications that use a categorical approach. Can you explain those implications of your framework?

Sarah Vendzules  32:13

Sure. Well, so the categorical approach and circumstance specific approach, those are all decisions made at the non discretionary step. So because of that, there's a lot more attention given to these decisions, a lot more litigation in this area, and the abuses aren't going to be as egregious as you're going to find at the discretionary phase. But I still think that the output portability framework is valuable here, because you do see the same types of logical errors that are being made in the discretionary context also being made at the non discretionary step. So, I might have touched on this briefly, but in immigration law, in determining whether a conviction is a ground of deportability or inadmissibility, or basically ineligibility for relief, the immigration adjudicators have to apply, in many cases, a categorical analysis. And what that means is they have to look at what facts have actually been found by the system. So what is the sort of least culpable behavior that a statute penalizes? And that's all we can assume that the immigrant actually did. Of course, nothing is ever simple in immigration law, so there are also circumstances where courts are able to apply what's called a modified categorical approach, and it's modified because the immigration adjudicator is allowed to look behind the criminal conviction to some degree by examining a certain set of documents, which are called record of conviction documents, and that includes the charging document, that includes the plea colloquy. And the purpose of the modified categorical approach is not to determine what really happened, not to make any fact finding, but just to determine what the person was actually convicted of, what crime the person was actually convicted of in cases where it's not clear because of the way a state statute is structured or worded, where there might be multiple possible crimes that the person could have been convicted of, and just having the certificate of disposition is not going to be enough: you need to just sort of dig a little deeper. And so under this modified approach, you're allowed to look at certain certain documents. My output portability framework, I think, helps to also understand the categorical approach in maybe a slightly different way than it's been understood, particularly with respect to the record of conviction documents, because you can consider them as outputs that have been adjudicated by the criminal legal system. So when you're looking at, for example, the charging document, if someone's been convicted of something, then they have been adjudicated guilty of that charge. Similarly, when you're looking at the plea colloquy, when they say that they when they are pleading guilty to something, they're admitting certain things under oath. All of that is just part of determining what has already been adjudicated by the criminal legal system. But because of this distinction between documents and outputs, I think there's been significant confusion in how we understand the categorical approach, or maybe it's more accurate to say slippage, because immigration is obviously always interested in in deporting more people rather than fewer. So when you have this list of documents that the Supreme Court has said, well, these are the record of conviction, a complaint might contain adjudicated and unadjudicated counts, a plea colloquy might contain statements that are not essential to the admission of the crime. They might contain, you know, other sort of factual narrative about the circumstances that led to the crime. And so, because the court is referring to these documents, sort of writ large, rather than to the outputs contained therein, the agency is taking it as permission to start looking at these other things, these other outputs that have not been adjudicated. And the only possible thing that they could be doing with these outputs is engaging, in fact, finding based on them. So, you know, if you see it that way, I think it is easier to understand that this is not what the court was talking about when it set out this categorical approach. The whole point is that the agency is not qualified to engage in fact finding, the agency is not qualified to second guess the criminal legal system and to determine what really happened in a case. And again, we're talking about you know, non discretionary determinations, sort of first step eligibility determinations, where it's pretty clear that there are rules and standards and the agency is supposed to follow them, and they're not supposed to deport somebody if they have not been convicted of this type of crime. So I think that the essential point is really that the output portability framework helps guard against any kind of slippage of the categorical approach by giving a different way to understand what is the record of conviction. Through the output portability lens, the record of conviction is really, or should be properly thought of, I would say, as criminal legal system outputs that are fully portable because they've already been adjudicated as true beyond a reasonable doubt. And you know, this would include the portion of the charging document that gives elements of the crime. It includes the facts that were admitted under oath during the plea colloquy or the written plea agreement, if there was one that were necessary to make out the crime. But it doesn't include this extraneous information that might happen to be contained in the same documents, but that was not adjudicated because it was not necessary to be to be adjudicated along the way that the conviction, and therefore there's no way this could possibly be helpful in discovering in a modified categorical analysis what crime the person was actually convicted of. So basically, the limiting principle of the categorical and the modified categorical approach is that adjudicators are limited to the facts that they were already found by the criminal legal system, and even more precisely, actually the facts that were necessarily found because they were essential elements of the crime of conviction. So the only way to really know if an output is part of the record of conviction is to determine if it's necessarily been adjudicated. And we should be looking at unadjudicated outputs that just happen to be found in the same document as basically garbage. They're not really part of the record of conviction, at least under my theory.

Peter Mason  38:15

Finally, what do you hope a listener or reader takes away from this episode more generally?

Sarah Vendzules  38:20

I would be happy if people come away with a sense of the basic unfairness of what is currently going on. I do feel like I've spent a good deal of my career sort of railing against this. And, you know, on the one hand, speaking to criminal defense attorneys who couldn't believe that's how these documents are being used, and on the other hand, speaking to immigration adjudicators and immigration judges, and for them, it's just, you know, part of a normal day to be to be doing this, to be accepting these outputs, or these documents, with basically no grain of salt whatsoever. And so I'm hoping that people would maybe take a second look, because I think that if we're actually being explicit about our choices, I don't think this is a choice that when you say it explicitly, I don't think it's a choice that we want to make. That anybody who's ever been arrested is now going to be essentially barred from getting any kind of immigration relief or any kind of discretionary immigration relief, which is almost every form of immigration relief. So to treat an arrest as if it were a conviction, I think that's wrong. And I think what I've tried to do is show that that is what is happening, and it is morally wrong, it is sort of prudentially wrong, it's not an explicit choice that we would make as a society. And then also, you know, trying to show why I think it's also legally wrong and why it can be challenged.

Peter Mason  39:45

Sarah Vendzules, thank you so much for joining us and discussing your article.

Sarah Vendzules  39:49

Thank you so much for having me. It's really been a pleasure.

Peter Mason  39:56

We hope you've enjoyed this episode of The California Law Review podcast. If you would like to read the article, you can find it in Volume 112, Issue Three of the California Law Review at californialareview.org. For updates on new episodes and articles, please follow us on Instagram @californialawreview. A complete list of our socials is available on our website. Lastly, you can find a list of the editors who worked on this episode of the podcast in the show notes. See you next time.