Ammons v. State

880 S.E.2d 544, 315 Ga. 149

Opinion

315 Ga. 149 FINAL COPY

S22A0542. AMMONS v. THE STATE.

BETHEL, Justice.

Mia Ammons is being prosecuted for driving under the influence of alcohol. She largely refused to cooperate when the state trooper who pulled her over sought to perform a preliminary breath test and various field sobriety tests, and she later refused to consent to a blood test for which no search warrant had been obtained by the police. She claims that any use of evidence of her refusal to perform the breath and field sobriety tests violates her right against self- incrimination under the Georgia Constitution. She similarly argues that two Georgia statutes that permit evidence of her refusal to consent to a blood test to be used against her violate the General Assembly’s constitutional duty to enact laws that protect Georgia citizens in the full enjoyment of their rights, privileges, and immunities as citizens.

The trial court denied Ammons’s motion to suppress evidence from the roadside stop, including her refusal to participate in a number of these tests, concluding that her constitutional arguments failed. We granted Ammons’s application for interlocutory review of the trial court’s decision.

As explained below, Ammons had the right to refuse to perform the preliminary breath test and the field sobriety tests under the Georgia Constitution, and evidence of her refusals cannot be introduced at her trial. We also determine that the Georgia Constitution’s Privileges and Immunities Clause does not bar the admission of evidence that she refused to consent to a blood test. We therefore affirm in part and reverse in part the trial court’s denial of Ammons’s motion to suppress.

1. Background.

Ammons was charged with driving under the influence (less safe) pursuant to OCGA § 40-6-391 (a) (1).1 She moved in limine to

1 OCGA § 40-6-391 (a) (1) provides that “[a] person shall not drive or be in actual physical control of any moving vehicle while . . . [u]nder the influence

3

suppress evidence from her roadside stop and her interactions with the trooper, including with regard to her refusal to consent to a preliminary breath test, field sobriety tests, and a blood test.

The record of the hearing on Ammons’s motion to suppress shows the following. Just after midnight on July 14, 2018, Ammons was driving her vehicle on a state highway in Paulding County when she was stopped by State Trooper Levi Perry because her car did not have a working light illuminating her license plate. After approaching Ammons’s car and smelling alcohol on her breath, Trooper Perry asked Ammons to step out of her car. Ammons did so. Trooper Perry testified that he “immediately noticed” that Ammons was “extremely unsteady.” In response to questions from Trooper Perry, Ammons said that she had consumed alcohol “a few hours prior” to the stop and that “she had a few beers.” Trooper Perry testified that, during their discussion, he noticed that Ammons had “bloodshot watery eyes,” seemed “withdrawn,” and had slurred

of alcohol to the extent that it is less safe for the person to drive . . . .” Ammons was also charged with a tag light violation (OCGA § 40-8-23) and with failure to change driver’s license address information (OCGA § 40-5-33).

4

speech.

As their conversation continued, Trooper Perry asked Ammons if she would provide a breath sample for a preliminary breath test. She refused. Trooper Perry then asked Ammons to stand with her back against his patrol car and asked her if she had any medical conditions. She replied that, other than needing to wear glasses, she did not. Trooper Perry then directed Ammons to “look straight at [him] and [to] follow the tip of [his] finger with her eyes only.” Ammons then did so for a brief period of time. Noting in his testimony that this was part of a horizontal gaze nystagmus (HGN) test, Trooper Perry testified that the test showed six out of six clues that Ammons was impaired.

Trooper Perry then began directing Ammons to perform a “walk and turn” test, but she refused to participate. Trooper Perry then arrested Ammons for DUI and read her the Georgia implied consent warning for suspects over the age of 21 and requested that

5

that she provide a blood sample.2 Ammons refused to answer when Trooper Perry asked her if she consented.

Trooper Perry testified that both a dashboard camera and a body camera he was wearing at the time recorded his interactions with Ammons. Both recordings were admitted at the hearing on the motion to suppress.

At the hearing, Trooper Perry testified that

[t]he purpose of the field sobriety and advanced roadside and impairment detection is to determine whether or not that person is indeed impaired to both give them the opportunity to counteract any initial suspicion and to . . .

determine at what level of impairment there is.

Trooper Perry testified that the standard battery of field sobriety tests begins with an assessment of the suspect’s medical conditions, such as recent head trauma or any problems with the suspect’s neck, back, or legs. Once it has been ascertained that no such conditions are present, an HGN test is performed, which involves an initial evaluation of “equal tracking” of the eyes between the “ten and two”

2 Ammons indicated to Trooper Perry that she was 30 years old at the time.

6

positions followed by three different evaluations: “detection of lack of smooth pursuit,” “sustained nystagmus at maximum deviation,” and “onset prior to maximum deviation.” These tests require the suspect to follow an object, such as the tip of the officer’s finger, with her eyes for several seconds. The HGN test evaluates whether there is “involuntary jerking of the eyes either caused by a medical condition or by impairment.”3 Trooper Perry testified that the HGN test requires the suspect’s participation and that “unless there’s cooperation you can’t perform it.” Following an HGN test, a suspect is then asked to perform a “walk and turn” test which is used to determine the suspect’s motor functions. The suspect is then typically asked to perform a “one-leg stand.”

Following the hearing, the trial court denied Ammons’s motion to suppress. Ammons moved for reconsideration, and the trial court entered an amended order denying the motion. In its order, the trial court determined that Ammons voluntarily performed the HGN test

3 Trooper Perry characterized the HGN test as the “most reliable portion”

of the standard battery of field sobriety tests.

7

and that the results of the test were not obtained in violation of her rights under the Georgia Constitution. The court also determined that, under our decision in Keenan v. State, 263 Ga. 569, 571-572 (2) (436 SE2d 475) (1993), Ammons’s refusal to perform the preliminary breath test could be admitted into evidence and that her refusal to perform field sobriety tests did not implicate her rights against self- incrimination under the Georgia Constitution because she was not in custody at the time of the refusal, citing Keenan and Long v. State, 271 Ga. App. 565, 567-569 (2) (610 SE2d 74) (2004). Finally, the trial court determined that, by allowing a defendant’s refusal to consent to a warrantless blood test as evidence of guilt in a criminal case, Georgia’s implied consent statutes, OCGA §§ 40-5-67.1 and 40-6- 392, do not violate the Privileges and Immunities Clause, the Due Process Clause, or the Search and Seizure Clause of the Georgia Constitution. The same day, the trial court issued a certificate of immediate review.

Ammons timely filed in this Court an application for interlocutory review, which we granted. We directed the parties to

8

address only the following questions:

1. Should this Court overrule its holding in [Keenan], that admission of evidence that a defendant refused a roadside alco-sensor test does not violate the Georgia Constitution’s guarantee of the right against compelled self-incrimination?

2. Does the Georgia Constitution’s guarantee of the right against compelled self-incrimination apply to field sobriety tests, such that evidence that the defendant refused to submit to such tests is inadmissible?

3. Do OCGA §§ 40-5-67.1 or 40-6-392 violate the Georgia Privileges and Immunities Clause?

Ammons timely appealed. We now address each of these questions in turn.

2. The Georgia Constitution’s protection against self-

incrimination applies to preliminary breath tests using an alco- sensor and field sobriety tests that require the cooperation of the suspect.

Article I, Section I, Paragraph XVI of the Georgia Constitution (“Paragraph XVI”) provides that “[n]o person shall be compelled to give testimony tending in any manner to be self-incriminating.” In Olevik v. State, 302 Ga. 228, 228 (806 SE2d 505) (2017), this Court held that this provision “applies to more than mere testimony; it also protects us from being forced to perform acts that generate

9

incriminating evidence.”4 Olevik specifically recognized that Paragraph XVI “prohibits law enforcement from compelling a person suspected of DUI to blow his deep lung air into a breathalyzer” for purposes of determining his blood alcohol content. Id. at 228-229. Two years later, in Elliott v. State, 305 Ga. 179, 210 (IV) (824 SE2d 265) (2019), we determined that admission of evidence that the defendant refused to consent to a chemical breath test likewise violates the rights protected by Paragraph XVI, noting that “Paragraph XVI generally prohibits admission of a defendant’s pretrial refusal to speak or act.” And earlier this year, we recognized that this protection extended to state-administered urine tests. See Awad v. State, 313 Ga. 99, 103 (3) (868 SE2d 219) (2022) (“Under Olevik and Elliott, the right against compelled self-incrimination protected by Paragraph XVI prohibits the State from admitting into

4 We went on to note in Olevik that

although Paragraph XVI refers only to testimony, its protection against compelled self-incrimination was long ago construed to also cover incriminating acts and, thus, is more extensive than the Supreme Court of the United States’s interpretation of the right against compelled self-incrimination guaranteed by the Fifth Amendment.

302 Ga. at 240 (2) (c) (ii).

10

evidence a defendant’s refusal to submit to a urine test when doing so would require a defendant to urinate into a collection container to generate a sample for chemical testing. This collection method necessarily requires a defendant to cooperate with the State by performing an act that generates self-incriminating evidence.”). In Awad, we noted that, like the chemical breath tests at issue in Olevik and Elliott, the urine test involved the State “asking the defendant to affirmatively give the State evidence from the defendant’s body in a particular manner that is neither natural nor automatic.” 313 Ga. at 103 (3).

(a) Admission of evidence that a defendant refused to provide a breath sample for a preliminary breath test using an alco-sensor violates the Georgia Constitution’s protection against self- incrimination.

More than two decades before we decided those cases, we suggested in Keenan that this constitutional protection did not apply to the type of preliminary breath test Ammons was asked to submit to in this case. In Keenan, the defendant, who was suspected of driving under the influence, refused to submit to a preliminary

11

breath test that would alert the officer to the presence of alcohol (what is sometimes also referred to as an “alco-sensor” test). See 263 Ga. at 569. In that case, over the defendant’s objection, the State was permitted to introduce evidence of his refusal to undergo the breath test. See id. at 571 (2). On appeal before this Court, the defendant argued that the Fifth Amendment to the United States Constitution barred introduction of evidence regarding his refusal. See id. This Court held that, because the defendant was not in custody at the time, Miranda warnings5 (which had not been given) were unnecessary, and “evidence of appellant’s refusal to undergo the alco-sensor test would not be inadmissible as violative of his constitutional right to remain silent.” Id. at 572 (2). After also determining that the admission of evidence regarding his refusal did not violate former OCGA § 24-9-20,6 this Court stated that “[t]here

5 See Miranda v. Arizona, 384 U. S. 436 (86 SCt 1602, 16 LE2d 694)

(1966).

6 Former OCGA § 24-9-20 provided that “[n]o person who is charged in

any criminal proceeding with the commission of any indictable offense or any offense punishable on summary conviction shall be compellable to give evidence for or against himself.” That Code section was repealed as part of the

12

was no violation of appellant’s right not to incriminate himself under the [F]ifth [A]mendment, the Georgia Constitution, or [former] OCGA § 24-9-20, because he was not in custody at the time the field sobriety test was requested.” (Punctuation omitted; emphasis supplied.) Id. (quoting Lankford v. State, 204 Ga. App. 405, 406 (2) (419 SE2d 498) (1992)).

That was the first and only mention of the Georgia Constitution in Keenan. Keenan pointed to no specific provision of the Georgia Constitution that was implicated by the issues in the case or that the appellant had argued was violated by the admission of evidence regarding his refusal to consent to the preliminary alco- sensor test. See generally id. And Keenan contained no analysis of any Georgia constitutional provision in support of its apparent holding. See generally id.

As noted above, though, in the years since Keenan was decided,

enactment of the current Evidence Code. However, OCGA § 24-5-506 (a) now provides that “[n]o person who is charged in any criminal proceeding with the commission of any criminal offense shall be compellable to give evidence for or against himself or herself.”

13

this Court has determined that Paragraph XVI offers a number of protections to a suspect who refuses to cooperate with police during a roadside DUI stop. Olevik recognized that Paragraph XVI protects a suspect from being compelled by the police to perform a chemical breath test that yielded a measurement of his blood alcohol content. See 302 Ga. at 246 (2) (c) (iv). Elliott determined that the suspect’s refusal to perform the test could not be used against him. See 305 Ga. at 223 (IV) (E). And Awad applied these same protections in the context of a urine test. See 313 Ga. at 106 (5).

Although we have never expressly overruled Keenan, it is clearly in tension with our holdings in Olevik, Elliott, and Awad, and we have already expressed doubts about Keenan’s seeming equation of the rights protected by former OCGA § 24-9-20 with those secured by Paragraph XVI and the soundness of that reasoning. See State v. Turnquest, 305 Ga. 758, 772 (4) (827 SE2d 865) (2019) (“We equated [former OCGA § 24-9-20] with Paragraph XVI without further analysis of the constitutional provision (which does not appear to have been raised by the appellant in that case) . . . .”). Moreover, the

14

Court of Appeals has recently applied Olevik and Elliott to determine that Paragraph XVI prohibits the State from admitting evidence of a defendant’s refusal to take the type of preliminary test Ammons refused here, even though the case before it involved “an alco-sensor preliminary breath test, rather than the type of [chemical] breathalyzer breath tests involved in Elliott and Olevik.” State v. Bradberry, 357 Ga. App. 60, 65-66 (3) (849 SE2d 790) (2020). The Court of Appeals determined that “[b]ecause [the defendant] had the right to refuse to provide incriminating evidence by performing such an affirmative act under Paragraph XVI, the admission of evidence of his refusal violates the state constitutional right against self-incrimination.” Id. at 66 (3).

Like the Court of Appeals in Bradberry, we see little distinction between the preliminary alco-sensor breath test Ammons refused to take during her roadside stop and the type of chemical breath tests at issue in Olevik and Elliott (or, for that matter, the urine test in Awad). Both a preliminary alco-sensor test and a chemical breath test require the defendant to affirmatively blow into a device “for a

15

sustained period of time.” Bradberry, 357 Ga. App. at 66 (3). And because the preliminary test detects the presence of alcohol, evidence generated by the test is plainly incriminating against a suspect who has consumed alcohol. See id. See also Olevik, 302 Ga. at 231 (1) (b) (noting that a portable alco-sensor test detects the presence of alcohol). We thus see little merit in the State’s efforts to distinguish that test from the ones considered in Olevik and Elliott.7 Moreover, stare decisis does not require us to perpetuate Keenan’s flawed holding.

Under the doctrine of stare decisis, courts generally stand by their prior decisions, because it promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process. Stare decisis, however, is not an

7 We also note that Keenan wrongly suggested that a suspect’s rights

under Paragraph XVI only come into force in a custodial setting. See 263 Ga. at 571 (2). As noted above, the affirmative act required by the alco-sensor test, just like the act required to perform the chemical breath tests discussed in Olevik and Elliott, plainly generates evidence against the suspect. Although the refusals at issue in Olevik, Elliott, and Awad all occurred post-arrest, nothing about the holdings of those cases or our consideration of the rights protected by Paragraph XVI suggested that those rights only come into force once the suspect is in custody. And our decision today makes clear that the rights guaranteed by Paragraph XVI protect a suspect from being compelled to perform an affirmative act that generates evidence against her, regardless of whether that act takes place before or after she is placed in custody.

16

inexorable command. Courts, like individuals, but with more caution and deliberation, must sometimes reconsider what has been already carefully considered, and rectify their own mistakes. In reconsidering our prior decisions, we must balance the importance of having the question decided against the importance of having it decided right. To that end, we have developed a test that considers the age of precedent, the reliance interests at stake, the workability of the decision, and, most importantly, the soundness of its reasoning. The soundness of a precedent’s reasoning is the most important factor. We have also said that stare decisis carries less weight when our prior precedent involved the interpretation of the Constitution, which is more difficult than statutory interpretation for the legislative process to correct. This doesn’t mean that we disregard stare decisis altogether, though; what it actually means is that the first stare decisis factor (soundness of reasoning) becomes even more critical. The more wrong a prior precedent got the Constitution, the less room there is for the other factors to preserve it.

(Citations, punctuation and emphasis omitted.) Olevik, 302 Ga. at 244-245 (2) (c) (iv).

As noted above, we see no plausible distinction between the breath test at issue in this case and those we considered in Olevik and Elliott, and we should not strain to find distinctions between Keenan and these more recent decisions where no meaningful ones exist. Moreover, the final sentence of the discussion in Keenan that

17

dealt with refusal to consent to an alco-sensor test was the first, and only, mention of the Georgia Constitution in that opinion. The opinion never quoted or referred to any specific provision of the Georgia Constitution, nor did it purport to tie its ruling to the text or history of any Georgia constitutional provision. In short, to the extent the statement in Keenan regarding the Georgia Constitution was even a holding, this Court offered no reasoning to support it. See Turnquest, 305 Ga. at 771 (4) (rejecting stare decisis as basis for upholding earlier decision where the Court’s “opinion did not cite, let alone analyze, any particular Georgia statute or Georgia constitutional provision in support of its holding”). Moreover, since that time, our Court has concluded “after extensive review of the historical record and our case law,” that Paragraph XVI prohibits the introduction of evidence of a defendant’s refusal to consent to a breathalyzer test in conjunction with a DUI stop. Elliott, 305 Ga. at 180. Thus, in light of our understanding and detailed explanation of what Paragraph XVI protects, the lone reference in Keenan to the Georgia Constitution was unsound, “which is the most important

18

stare decisis consideration, especially in constitutional cases.” Turnquest, 305 Ga. at 773 (4).

In addition, “[n]one of the remaining stare decisis factors indicate that we should retain this unfounded decision.” Turnquest, 305 Ga. at 774 (4). Keenan was decided 29 years ago, and we have overruled decisions older than that. See Southall v. State, 300 Ga. 462, 468 (1) (796 SE2d 261) (2017) (overruling a 45-year-old precedent on premature motions for new trial); State v. Hudson, 293 Ga. 656, 661-662 (748 SE2d 910) (2013) (overruling a 38-year-old precedent regarding when a new post-appeal sentence is unconstitutionally vindictive); State v. Jackson, 287 Ga. 646, 659- 660 (5), (6) (697 SE2d 757) (2010) (overruling a nearly 29-year-old interpretation of the felony murder statute). Keenan also created none of the reliance interests of the type normally given weight in stare decisis analysis, namely those relating to property or contractual rights, and any reliance interests that may have developed around the practice of introducing evidence of a suspect’s refusal to perform the test

19

do not outweigh the countervailing interest that all individuals share in having their constitutional rights fully protected. If it is clear that a practice is unlawful, individuals’ interest in its discontinuance clearly outweighs any law enforcement entitlement to its persistence. The mere fact that law enforcement may be made more efficient can never by itself justify disregard of constitutional rights.

(Citation and punctuation omitted.) Olevik, 302 Ga. at 246 (2) (c) (iv). Finally, “[t]he remaining factor of workability is not reason enough to preserve” Keenan. Id. As we discussed in Olevik with regard to chemical breath tests,

law enforcement may have to consider whether a suspect has validly waived his right against self-incrimination under the totality of the circumstances. We recognize that requiring this determination before administering a [preliminary] breath test [using an alco-sensor] is more difficult than simply waiting for an affirmative response

to an officer’s request to perform the test. Id. “But this difficulty is not reason enough to persist” in Keenan’s error. Id.

Consequently, to the extent Keenan purported to issue a holding on the issues in that case pursuant to the Georgia Constitution, any such holding is overruled. And because the trial court’s order denying Ammons’s motion to suppress relied in part on

20

Keenan, we reverse that portion of the trial court’s ruling.8

(b) The protections of Paragraph XVI apply to field sobriety tests that require the suspect’s cooperation.

We also answer in the affirmative the second question posed in this appeal: that is, whether the Georgia Constitution’s guarantee of the right against compelled self-incrimination applies to field sobriety tests that require the suspect’s cooperation, such that evidence that the defendant refused to submit to such tests is inadmissible. We therefore reverse the trial court’s rulings in regard to Ammons’s refusal to participate in some of the field sobriety tests

8 We note that the decision of the Court of Appeals in Bradberry did not

cite or discuss Keenan and instead simply applied Elliott and Olevik. We also note that the trial court’s order denying Ammons’s motion to suppress examined this issue with both Keenan and Bradberry in mind. The trial court rightly determined at the time that, in light of the fact that Keenan had never been overruled by this Court, Keenan controlled even though Bradberry followed our more recent line of precedents, including Olevik and Elliott. See Ga. Const. of 1983, Art. VI, Sec. VI, Par. VI (“The decisions of the Supreme Court shall bind all other courts as precedents.”). See also State v. Stanford, 312 Ga. 707, 710 n.3 (864 SE2d 448) (2021) (noting “the general rule that lower courts must follow this Court’s precedent until we overrule it”). Although Olevik and Elliott suggested as much, we commend the trial court in adhering to binding authority and resisting the temptation afforded by Bradberry. With Keenan as the controlling authority on the question, its holding could only be dislodged by this Court, and this opinion should at last clarify that Keenan is no longer good law as to the issue of whether Paragraph XVI allows a suspect to refuse to consent to the preliminary breath test and protects that suspect from having his refusal used against him at trial.

21

Trooper Perry attempted to perform during the roadside stop.

As with other tests performed by the police during DUI stops, a field sobriety test is designed to “reveal some other condition or impairment” of the driver relevant to determining whether he or she was driving under the influence. Mitchell v. State, 301 Ga. 563, 570- 571 (3) (802 SE2d 217) (2017), disapproved on other grounds by Turnquest, 305 Ga. at 775 (4) n.15. As we noted in Mitchell, “field sobriety tests may involve specific, unusual maneuvers that are . . . intended to reveal . . . [s]uch characteristics as unsteady gait, lack of balance and coordination, impaired speech, lack of memory, or inability to divide one’s attention. . . . .” 301 Ga. at 571 (3). Although field sobriety tests are not a search within the meaning of the Fourth Amendment, see id., such tests are clearly designed to generate incriminating evidence against a person suspected of driving under the influence.

Moreover, it is clear that the suspect’s cooperation is required in order to perform the “specific, unusual maneuvers” characteristic of the standard field sobriety tests Trooper Perry sought to perform

22

here. At the hearing on Ammons’s motion to suppress, Trooper Perry testified that “unless there’s cooperation you can’t perform [an HGN test].” Moreover, in its brief, the State notes that

[t]he [walk and turn] and [one-leg stand] test[s] and other dexterity test[s] that cause a DUI suspect to divide [her]

attention among more than one task, also known as divided attention tests, enable an officer to better know whether that suspect is impaired by an intoxicant to the point of being a less safe driver. These divided attention tests are revealing as to impairment because a motor vehicle driver is required to divide [her] attention among several tasks at once while driving.

As the Court of Appeals has discussed at length, like the HGN test, both the walk and turn test and the one-leg stand test plainly require the suspect to cooperate by performing affirmative acts. See, e.g., Davis v. State, 358 Ga. App. 832, 835 (856 SE2d 411) (2021) (noting that, at the request of the officer, the suspect “performed” the walk and turn test and the one-leg stand); Leggett v. State, 354 Ga. App. 877, 878 (842 SE2d 313) (2020) (noting that the suspect “could not keep his balance” while performing the walk and turn test); State v. Culler, 351 Ga. App. 19, 21 (830 SE2d 434) (2019) (noting that the suspect “was able to stand straight, arms at his side,

23

with one leg raised, and while counting out loud for approximately 23 seconds, during which time he exhibited no problems with his balance, coordination, or speech” and “ceased performing the test only when [the officer] directed him to do so”); Oh v. State, 345 Ga. App. 729, 730-731 (815 SE2d 95) (2018) (noting that the officer instructed the suspect to perform the walk and turn test by instructing him to “take nine steps with his arms to his side and then turn around” and that signs of impairment included taking an incorrect number of steps, failing to maintain balance, making an improper turn, and using arms to maintain balance). And such tests cannot be performed if the suspect is not in a condition to cooperate. See Olevik, 302 Ga. at 231 (1) (b) (noting that the “walk-and-turn and one-leg-stand tests were not conducted because [the suspect] had certain physical limitations”); Adams v. State, 344 Ga. App. 159, 168 (4) (809 SE2d 87) (2017) (noting that the officer “chose not to perform the walk-and-turn and one-leg stand field sobriety tests because [the suspect] was too unsteady on his feet”); Miller v. State, 343 Ga. App. 197, 197 (806 SE2d 648) (2017) (noting that because

24

the suspect informed the officer that “she had hip issues,” the officer “determined that [the suspect] could not be medically cleared to perform the one-leg-stand or walk-and-turn evaluations”).

The State suggests that this information is merely useful to the officer in establishing probable cause for an arrest, but information that is useful for that purpose is also clearly useful to the State in proving at trial that the defendant violated the DUI statute. Moreover, while the level of cooperation for each standard field sobriety test appears to be somewhat different from the chemical breath tests at issue in Olevik and Elliott and the urine test in Awad, it is clear that field sobriety tests that require the suspect to cooperate by performing some affirmative act are covered by the protections of Paragraph XVI. See Awad, 313 Ga. at 103 (3) (applying Paragraph XVI to state-administered urine tests and noting that such tests involved asking the defendant to perform a task which was “neither natural nor automatic”); Olevik, 302 Ga. at 243-244 (2) (c) (ii) (applying Paragraph XVI to breath tests because, among other reasons, “it is required that the defendant cooperate by

25

performing an act” (emphasis omitted)). Ammons had the right to refuse to engage in these tests, and, except with regard to the HGN test, she did refuse. Her refusal to perform the remaining field sobriety tests cannot be used against her at trial. See Awad, 313 Ga. at 106 (5); Elliott, 305 Ga. at 223 (IV) (E); Olevik, 302 Ga. at 246 (2) (c) (iv). Accordingly, we reverse that portion of the trial court’s order that reached a contrary result.

(c) We decline the State’s invitation to reconsider Olevik, Elliott, and Awad and reject the interpretive principles advanced in the dissent.

Both the State and the dissent argue that the entire basis of our holdings in Olevik, Elliott, and Awad was flawed and should be reconsidered in this case. But we see no reason to do so.

First, the dissent rejects long-standing interpretive principles and replaces them with a too-narrow focus on isolated words divorced from history and context. This novel approach would ignore all of our case law on constitutional interpretation before 1983 and begin anew with 1983 dictionary definitions. This simply is not how we have ever engaged in constitutional interpretation in Georgia.

26

See, e.g., Elliott, 305 Ga. at 184-187 (II) (B) (detailing historic interpretive approach involving consideration of judicial construction of previous constitutions).

Second, no reasonable observer during the drafting and ratification of the 1983 Constitution would have understood the provisions of the proposed new constitution to be understood without reference to the construction of their predecessors. See Select Committee on Constitutional Revisions, 1977-1981 (“Select Committee”), Transcript of Meetings, Committee to Revise Article I, meeting of Subcommittee to Revise Section I, Oct. 4, 1979, p. 69 (noting that the search and seizure clause had “been construed so many times” and a “tremendous body of law” developed on the words of that clause that “if we change much of that we’re going to open a complete new field”); id. at 97 (noting that the committee would “open up a keg of worms” if it “monkey[ed] with” the double jeopardy clause); id. at 103-106 (notwithstanding members’ uncertainty about meaning of phrase “corruption of blood,” Justice Bowles noted that the phrase had been defined in case law, and another committee

27

member suggested the phrase remain in the light of that case law); id. at 51, meeting of Subcommittee on Rights of Persons, Oct. 25, 1979 (Justice Bowles noted “change should be made where change is necessary but” courts view a change in words as “an intention on the part of the framers to give it a different meaning from the meaning that theretofore existed”); id. at 22-29, Nov. 9, 1979, meeting of Full Committee (one committee member proposed removing the word “remonstrance” from provision on right to assemble and petition, but majority of committee voted to keep the provision as written after argument was made that the alternative language omitting the word would narrow the right) (cited in Elliott, 305 Ga. at 208-209 (III) (C) (ii)). And the dissent’s proposed new theory would upend any number of critical legal issues that have long been understood as well settled through application of the interpretive principles summarized in Elliott. See, e.g., Thompson v. Talmadge, 201 Ga. 867, 885 (2) (41 SE2d 883) (1947) (resolving Three Governors Controversy in part through application of prior construction canon).

28

Third, for its principal case law support, the dissent relies almost exclusively on Drake v. State, 75 Ga. 413 (1885), and its purported conflict with Day v. State, 63 Ga. 667 (1879). In doing so, the dissent brushes past the fact that we rejected the dissent’s reading of Drake in 1889 and then again in 1916, and Drake has never again been cited for the dissent’s proposition. See Calhoun v. State, 144 Ga. 679, 680 (87 SE 893) (1916) (rejecting language on which the dissent relies as dicta and that “an examination of the facts of the case will show that the actual ruling was that the constitutional privilege does not prevent the introduction in evidence or the exhibition to the jury of clothing or any other article taken from a person accused of crime, where they tend to show his guilt”); Evans v. State, 106 Ga. 519, 521 (32 SE 659) (1899) (“[A]n examination of the facts appearing of record in [Drake] will show that it is really not in conflict with the Day case.”).9

9 As an extension of its rejection of Day, the dissent, in footnote 34, also

calls for this Court to overrule or disapprove 36 cases that followed Day. Of particular note, this list of cases cited by the dissent includes at least one decision from each decade, beginning in the 1870s and continuing to the 2020s.

29

Fourth, critical to the dissent is its presumption that our historical construction of our constitutional protection against compelled self-incrimination was wrong at the outset, citing one line from Olevik that if we were construing that provision “in the first instance, we might conclude” that the Georgia right was the same as the federal right. 302 Ga. at 235 (2) (c). This ignores our more extended treatment of the question in Elliott, which — while stopping short of determining “conclusively that Day was correctly decided,” 305 Ga. at 209 — outlined substantial evidence that Day’s holding was consistent with the original public meaning of the provision when it was adopted in 1877. See Elliott, 305 Ga. at 195- 202 (III) (B). The dissent fails to engage with any of that analysis.

Finally, the remaining arguments that the dissent puts forth were all considered at length and unanimously rejected in Elliott (many of which had already been previously considered and

This chain represents a long-standing and consistently applied body of case law regarding our state Constitution, and the dissent engages in no stare decisis analysis with regard to whether this Court should discard that line of decisions even if it began in error.

30

unanimously rejected in Olevik). Because they are not based on any previously unaddressed theory and do not point to any previously unconsidered precedent, we see no reason whatsoever to reconsider them yet again, despite the State’s invitation to do so.

3. Ammons has not met her burden to establish that the implied consent statutes violate Article I, Section I, Paragraph VII of the Georgia Constitution of 1983.

Finally, Ammons contends that, by allowing her refusal to consent to a blood test to be introduced as evidence at her trial, Georgia’s implied consent statutes, OCGA §§ 40-5-67.1 (b) and 40-6- 392, violate Article I, Section I, Paragraph VII of the Georgia Constitution of 1983 (“Paragraph VII”).10 As we understand it, her theory is that she invoked her right under the Georgia Constitution’s

10 Article I, Section I, Paragraph VII of the Georgia Constitution of 1983

provides that “[a]ll citizens of the United States, resident in this state, are hereby declared citizens of this state; and it shall be the duty of the General Assembly to enact such laws as will protect them in the full enjoyment of the rights, privileges, and immunities due to such citizenship.”

31

Search and Seizure Clause11 and Due Process Clause12 to insist that the police obtain a search warrant or satisfy some other exception to the warrant requirement before performing the test. And although our cases construing these provisions do not hold or suggest that a suspect’s refusal to consent to a blood test cannot be used against her at trial, she says Paragraph VII prohibits such use, because it imposes a “duty” on the General Assembly to enact laws that will protect citizens “in the full enjoyment of the rights, privileges, and immunities.” (Emphasis supplied.) Citing dictionary definitions (and little else), she claims that this language prohibits the General Assembly from imposing any degree of “burden” on her constitutional rights. In other words, she reads Paragraph VII to

11 Article I, Section I, Paragraph XIII of the Georgia Constitution provides that “[t]he right of the people to be secure in their persons, houses, papers, and effects against unreasonable searches and seizures shall not be violated; and no warrant shall issue except upon probable cause supported by oath or affirmation particularly describing the place or places to be searched and the persons or things to be seized.”

12 Article I, Section I, Paragraph I of the Georgia Constitution provides

that “[n]o person shall be deprived of life, liberty, or property except by due process of law.”

32

add a significant measure of extra or prophylactic protection of rights beyond what the provisions recognizing those rights cover.

We reject this claim. As an initial matter, Ammons’s burden to establish this claim is a difficult one. We presume that statutes are constitutional, and before an act of the General Assembly can be declared unconstitutional, “the conflict between it and the fundamental law must be clear and palpable and this Court must be clearly satisfied of its unconstitutionality.” (Citation and punctuation omitted.) S&S Towing & Recovery v. Charnota, 309 Ga. 117, 118 (1) (844 SE2d 730) (2020). “Because all presumptions are in favor of the constitutionality of a statute, the burden is on the party claiming that the law is unconstitutional to prove it.” (Citation and punctuation omitted.) Id. at 119 (1). And Ammons’s task is made all the more difficult because, to make this argument, she is asserting a novel and quite expansive construction of a provision of the Georgia Constitution that has received little attention since it was enacted.

33

Ammons has not made even the prima facie showing that would meet her heavy burden. Construing a constitutional provision, especially as an original matter, requires careful attention to not only the language of the clause in question, but also its broader legal and historical context, which are the primary determinants of a text’s meaning. See, e.g., Olevik, 302 Ga. at 235- 236 (2) (c) (i) (“We interpret a constitutional provision according to the original public meaning of its text,” for which we consider both the text’s “plain and ordinary meaning” and “the broader context in which that text was enacted” (citation and punctuation omitted)). This kind of analysis is especially difficult when the language in question was first enacted long ago and rarely interpreted since, because those important contextual clues can be more difficult to unearth, and the ordinary meaning of language can change over time. But Ammons’s showing with respect to the meaning of Paragraph VII grapples with none of this difficult analysis. Instead, she plucks isolated text from the constitutional provision, cites a single dictionary, and relies on general statements from a handful

34

of our decisions that do not interpret the relevant constitutional language. This kind of analysis does not meet the burden required to establish that Paragraph VII has the expansive reach that Ammons would have us recognize.

Given Ammons’s failure to meet her burden here, we need not reach any definitive conclusions as to the scope of Paragraph VII. But we do think it is helpful to provide some explanation of why Ammons’s claim, as articulated here, fails from the start. A general review of the legal and historical context relevant to Paragraph VII helps illustrate the deficiency of Ammons’s theory and offers no meaningful support for her novel and expansive view of the Georgia Constitution’s Privileges and Immunities Clause.

In considering the meaning of this clause, we begin by outlining some principles of constitutional interpretation. As a starting point,

we interpret the Georgia Constitution according to its original public meaning. And, of course, the Georgia Constitution that we interpret today is the Constitution of 1983; the original public meaning of that Constitution is the public meaning it had at the time of its ratification

35

in 1982.

Elliott, 305 Ga. at 181 (II). However, where a provision of the current constitution has been carried forward from a previous constitution, “we generally presume that a constitutional provision retained from a previous constitution without material change has retained the original public meaning that provision had at the time it first entered a Georgia Constitution, absent some indication to the contrary.” Id. at 183 (II) (A). See also Lathrop v. Deal, 301 Ga. 408, 428-432 (III) (B) (801 SE2d 867) (2017) (interpreting Article I, Section II, Paragraph V of the Constitution of 1983 in the light of the original public meaning of the provision as it first appeared in the Constitution of 1861).

Paragraph VII finds its roots in the period immediately after the Civil War. In 1868, to satisfy the conditions set by Congress for readmission to the Union, Georgia ratified a new constitution. See Macon and Augusta R. Co. v. Little, 45 Ga. 370, 374-375 (1) (1872) (noting that formation of a new state constitution and approval of that constitution by Congress were conditions for Georgia’s

36

reinstatement to the Union). As directed by Congress, that new constitution had to do two things: “conform[ ] with the Constitution of the United States in all respects”; and ensure “that the elective franchise shall be enjoyed by all persons [male and at least 21 years old]” of “whatever race, color, or previous condition.” See First Reconstruction Act of 1867, § 5 (1867). The resulting Georgia Constitution of 1868 included the predecessor to Paragraph VII, which read in full:

All persons born, or naturalized, in the United States, and resident in this State, are hereby declared citizens of this State, and no laws shall be made or enforced which shall abridge the privileges or immunities of citizens of the United States, or of this State, or deny to any person within its jurisdiction the equal protection of its laws. And it shall be the duty of the General Assembly, by appropriate legislation, to protect every person in the due enjoyment of the rights, privileges[,] and immunities guaranteed in this Section.

Ga. Const. of 1868, Art. I, Sec. I, Par. II.13

13 We acknowledge that (1) Paragraph VII’s language has changed since

1868, and (2) its affirmative language making it “the duty of the General Assembly to enact such laws as will protect [citizens] in the full enjoyment of the rights, privileges, and immunities due to such citizenship” (which remains similar to the 1868 predecessor of Paragraph VII) is unique, given that it places

37

One piece of context important for understanding the meaning of this provision is Section 1 of the Fourteenth Amendment to the United States Constitution.14 That provision, in relevant part, is materially the same as the first two sentences of the predecessor to Paragraph VII.15 The United States Supreme Court has construed

an affirmative duty on the legislature rather than restricting the government from taking certain actions, as, for instance, the language of Section 1 of the Fourteenth Amendment does. However, Ammons has pointed to nothing suggesting that the changes in Paragraph VII’s wording are material to the question before us in this case or that the language creating a duty on the part of the General Assembly requires suppression of evidence of her refusal to consent to a warrantless search.

14 We have recognized that “[w]hen interpreting a provision of our

Constitution that parallels a provision of the United States Constitution, we should take seriously decisions of the United States Supreme Court that have interpreted that parallel provision.” Elliott, 305 Ga. at 187 (III) (C). “But we owe those federal decisions no obedience when interpreting our own Constitution.” Id. And “any decision about the scope of a provision of the Georgia Constitution must be ‘rooted in the language, history, and context’ of that provision.” Id. (quoting Olevik, 302 Ga. at 234 (2) (b) n.3).

15 That language reads as follows: “All persons born or naturalized in the

United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside. No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States . . . .”

This Court has recognized that Section 1 of the Fourteenth Amendment and the first sentence of Article I, Section I, Paragraph II of the 1868 Georgia Constitution are “in substance, . . . identical.” White v. Clements, 39 Ga. 232, 269 (7) (1869) (Brown, C. J., concurring). See also Eric Foner, The Second Founding: How the Civil War and Reconstruction Remade the Constitution 90 (2019) (noting that “Georgia copied the wording of Section 1 [of the Fourteenth Amendment] into its state constitution”).

38

this Privileges or Immunities Clause of the Fourteenth Amendment as a guarantee to all people born or naturalized in the United States, including those recently freed from slavery, of citizenship and a collection of rights (the “privileges or immunities”) attributable to that status. See McDonald v. City of Chicago, 561 U. S. 742, 808 (130 SCt 3020, 177 LE2d 894) (2010) (Thomas, J., concurring). See also Strauder v. West Virginia, 100 U. S. 303, 306 (25 LE 664) (1879) (explaining that each of the provisions of Section 1 of the Fourteenth Amendment had a “common purpose”: “securing to a race recently emancipated, a race that through many generations had been held in slavery, all the civil rights” that were enjoyed by white citizens); Slaughter-House Cases, 83 U. S. 36, 100-101 (21 LE 394) (1872) (“If under [Article IV of the United States Constitution] equality of privileges and immunities is secured between citizens of different States, under the [F]ourteenth [A]mendment the same equality is secured between citizens of the United States.”).

The United States Supreme Court has construed the collection of rights protected by the Privileges or Immunities Clause of the

39

Fourteenth Amendment quite narrowly: it has said that it “protects only those rights ‘which owe their existence to the Federal government, its National character, its Constitution, or its laws,’” and “that other fundamental rights — rights that predated the creation of the Federal Government and that ‘the State governments were created to establish and secure’ — were not protected.” McDonald, 561 U. S. at 754 (quoting Slaughter-House Cases, 83 U. S. at 76, 79). Many judges and legal scholars have criticized this narrow construction of the Fourteenth Amendment’s Privileges or Immunities Clause, but those critics contend that the clause provided for “federal enforcement of constitutionally enumerated rights against the States,” not just a prohibition against “state- sponsored discrimination.” See id. at 840-841 (Thomas, J., concurring). See also Eric Foner, The Second Founding: How the Civil War and Reconstruction Remade the Constitution 73-76 (2019) (arguing that the clause was understood by some proponents of the Fourteenth Amendment to be the vehicle through which the rights guaranteed by the Bill of Rights would be applied to the States).

40

Ammons’s argument that Paragraph VII not only protects those rights enumerated in the United States or Georgia Constitutions, but also enhances or adds to those rights to some significant degree, is of a different character entirely and finds no support in either the Fourteenth Amendment’s Privileges or Immunities Clause or in the debate surrounding its crafting.

Nor does Ammons’s novel construction find support in this Court’s own contemporaneous construction of the 1868 Constitution’s predecessor to Paragraph VII. In White v. Clements, 39 Ga. 232 (1869), right after the 1868 provision was adopted, this Court was asked to determine whether a man who had won an election for public office was ineligible to serve because he was one- eighth black. Id. at 240. This Court held that the man was eligible because the predecessor to Paragraph VII made it clear he was a citizen of Georgia. Id. at 263-264 (7).16 Pertinent here, we explained

16 The declaration of citizenship for all residents of the state was not idly

made. There were lingering questions after the war about the legal status of people who were formerly enslaved. See, e.g., Foner at 55 (noting the “profound, difficult questions arising from the Civil War and the destruction of

41

that adopting the predecessor to Paragraph VII meant that the formerly enslaved “are citizens, and ‘citizens’ of this State. . . . This section of the Constitution of 1868[ ] takes another step — they become citizens — they grant to themselves the character of citizens.” (Emphasis in original.) Id. at 259 (7). See also id. at 273 (Brown, C. J., concurring) (“Whatever may or may not be the privileges and immunities guaranteed to the colored race, by the Constitution of the United States and of this State, it cannot be questioned that both Constitutions make them citizens” (emphasis omitted)). As an early construction of the predecessor to Paragraph VII, White is a good indication of the Clause’s original public meaning, and it does not support Ammons’s reading.

From the context set out above, we can surmise that the predecessor to Paragraph VII was understood as having an important role in guaranteeing that those who had been recently freed from slavery were citizens of Georgia and entitled to the same

slavery,” including “what rights should the former slaves enjoy and who should enforce them?”).

42

rights as other citizens. Even so, much like the scope of the “privileges or immunities” protected by the Fourteenth Amendment is subject to debate, the scope of the “rights, privileges, and immunities” protected by Paragraph VII is not entirely clear. But nothing that we have seen so far suggests that Paragraph VII does more than guarantee existing, enumerated rights to all citizens of the United States who reside in Georgia. We do not rule out the possibility that Paragraph VII does something more than that, but Ammons has not made that showing here. Her claim that Paragraph VII requires the suppression of evidence of a refusal to consent to a warrantless search therefore fails.17

Judgment affirmed in part and reversed in part. All the Justices concur, except McMillian and Colvin, JJ., who concur in part and dissent in part.

17 By holding that Ammons has not carried her burden to show that Paragraph VII requires the suppression of evidence of a refusal to consent to a warrantless search, we do not decide that such evidence is necessarily admissible in every case. Our rules of evidence or other applicable laws may result in the exclusion of such evidence in a given case. See, e.g., OCGA § 24- 4-403 (“Relevant evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.”).

43

ELLINGTON, Justice, concurring.

I join fully in the majority opinion and write separately only to emphasize the limitations of our holding in Division 3. As the majority notes, although Ammons has not marshaled authorities sufficient to persuade us that Paragraph VII does more than guarantee existing, enumerated rights to all citizens of the United States who reside in Georgia, we are not ruling out the possibility that Paragraph VII does do more.

Our jurisprudence on the meaning of the unique clause in Paragraph VII — “and it shall be the duty of the General Assembly to enact such laws as will protect them in the full enjoyment of the rights, privileges, and immunities due to such [Georgia] citizenship” — is scant. But the people of Georgia saw fit to include this clause in our Constitution, so we cannot brush it aside. And we have in broad terms recognized Paragraph VII as the source of “the General Assembly’s affirmative constitutional duty” to protect “the right of the people to exercise their civil rights[.]” State v. Miller, 260 Ga. 669, 672 (1) (398 SE2d 547) (1990). Future cases may present the

44

opportunity to develop a deeper understanding of the meaning of “the full enjoyment of the rights, privileges, and immunities” that citizens enjoy, as well as a better understanding of the affirmative duty imposed on the General Assembly to protect that enjoyment. In the present case, however, Ammons has not met her heavy burden of overcoming the presumption that the statutory evidentiary rule regarding blood test refusals in DUI cases is constitutional, so we must reject her Paragraph VII challenge to OCGA §§ 40-5-67.1 (b) and 40-6-392 (d).

45

PINSON, Justice, concurring.

I concur in the majority opinion, including its faithful application of this Court’s recent decisions in Olevik v. State, 302 Ga. 228 (806 SE2d 505) (2017), Elliott v. State, 305 Ga. 179 (824 SE2d 265) (2019), and Awad v. State, 313 Ga. 99 (868 SE2d 219) (2022). Perhaps there is room for debate about whether Olevik and Elliott were correct that the right against compelled self-incrimination under the Georgia Constitution protects affirmative acts. But those decisions plainly control here, and I am quite certain that stare decisis requires us to follow them. I write separately to explain why.

1. When courts consider whether to adhere to past decisions, stare decisis is the strong default rule. Some of the reasons for this rule are practical: applying stare decisis makes a body of law more stable, predictable, and reliable, and it deters the inefficient and expensive “endless relitigation” of basic and settled legal rules. Kimble v. Marvel Entm’t, LLC, 576 U.S. 446, 455 (135 SCt 2401, 192 LE2d 463) (2015). See also Cobb v. State, 187 Ga. 448, 452 (200 SE 796) (1939) (“The application of the doctrine of stare decisis is

46

essential to the performance of a well-ordered system of jurisprudence.”). But in my view, stare decisis is rooted most securely in the rule of law. See State v. Jackson, 287 Ga. 646, 658 (5) (697 SE2d 757) (2010). In our constitutional structure, courts have the special duty to say what the law is (as needed to resolve the controversies that come before us). Once we have decided a disputed issue of law, following that decision in future cases—treating like cases alike—promotes a system of equal treatment under the law rather than one built on “arbitrary discretion.” The Federalist No. 78, at 529 (Alexander Hamilton) (Jacob E. Cooke ed., 1961). And at a more fundamental level, following a past decision confirms that it is law, and that even the Court, like any other government actor and the litigants before us, is bound by it. This is the essence of the rule of law, and each time we overrule a past decision—choosing not to follow what our Court has said the law is—we risk chipping away at its foundation.

Of course, sometimes that’s a risk we must take. Even knowing the potential cost to the rule of law, courts in every jurisdiction

47

across our country, including this Court, have overruled plenty of past decisions. That’s because the rule of law can also be undermined by perpetuating decisions that are obviously and harmfully wrong. When sticking to such decisions would cause more damage to the rule of law than correcting course, courts may choose overruling as the lesser evil. See, e.g., Ellison v. Georgia R. Co., 87 Ga. 691, 696 (1) (13 SE 809) (1891) (Bleckley, C. J.) (when encountering “a great and glaring error affecting the current administration of justice,” the “maxim for a supreme court . . . is not [s]tare decisis, but [f]iat justitia ruat coelum [let justice be done, though the heavens fall]”).

It is not always easy to figure out when a past error is damaging enough to the rule of law that overruling the decision is worth the cost. Compare Cook v. State, 313 Ga. 471, 484-506 (3) (870 SE2d 758) (2022) with id. at 508-20 (Peterson, J., dissenting). But one important threshold question is whether the past decision can reasonably be understood as doing law. Was the decision “deliberate[ ],” the product of applying sound and accepted legal

48

principles to reach a reasoned answer to a disputed question, or was it a “hasty and crude” decision that seems conclusory, arbitrary, or based on something other than law, like personal preference? Doe v. Roe, 23 Ga. 82, 87 (1857), overruled on other grounds by Gresham v. Webb, 29 Ga. 320 (1859).

The first kind of decision—call it the “deliberate” kind—

ordinarily poses little threat to the rule of law, even when it is arguably wrong in hindsight. Any number of disputed legal questions are subject to reasonable debate. When one of those questions is presented to a court, the court’s constitutional role is to resolve it. If a court reaches that resolution through a decision that carefully applies sound, generally accepted legal principles, 18 it is

18 For example, when a court must address a question of constitutional

or statutory interpretation, we would hope to see some attention paid to the language and context of that provision to figure out what it meant at the time it was enacted. See, e.g., McIver v. State, 314 Ga. 109, 116 (2) (875 SE2d 810) (2022) (considering the “extensive history” of the law of involuntary manslaughter, including the “structure and history of the [statutory] text and the broader context in which [it] was enacted, including statutory and decisional law” (citation and punctuation omitted)); Seals v. State, 311 Ga. 739, 740 (1) (860 SE2d 419) (2021) (reviewing “the structure and history of the text and [its] broader context,” including “statutory and decisional law,” to determine the “ordinary meaning” of statutory language governing whether a

49

clear evidence of a proper exercise of the judicial power—that is, that the court is simply doing the job our Constitution gives it.19 See Judicial Council of Ga. v. Brown & Gallo, LLC, 288 Ga. 294, 297 (702 SE2d 894) (2010) (“The judicial power is that which declares what law is, and applies it to past transactions and existing cases; it expounds and judicially administers the law.” (cleaned up; emphasis in original) (quoting Thompson v. Talmadge, 201 Ga. 867, 874 (1) (41 SE2d 883) (1947))). Thus, recognizing such a decision as law and following it, even in the face of doubt about whether it was correct as an original matter, is consistent with the rule of law. See Patterson v. State, 299 Ga. 491, 516 (4) (789 SE2d 175) (2016)

case is final and appealable (citations and punctuation omitted)).

19 The judicial power has long been understood to include the power to

“liquidate,” or settle with finality, disputes about the meaning and operation of written laws. The Federalist No. 37, at 236 (James Madison) (Jacob E. Cooke ed., 1961) (“All new laws, though penned with the greatest technical skill, and passed on the fullest and most mature deliberation, are considered as more or less obscure and equivocal, until their meaning be liquidated and ascertained by a series of particular discussions and adjudications.”). See also The Federalist No. 22, at 143-44 (explaining that “[l]aws are a dead letter without courts to expound and define their true meaning and operation,” and advocating for a supreme court “to settle and declare in the last resort, an uniform rule of civil justice” because “[t]here are endless diversities in the opinions of men”).

50

(Blackwell, J., dissenting) (arguing that when we “get it wrong” on a question about the meaning of a statute, “it may be more appropriately left to the General Assembly to set things right” so long as “we have made our best effort” to apply “familiar and settled principles of statutory interpretation”). In my view, if nothing has changed besides the makeup of the court, overruling those kinds of decisions merely because the new personnel would come out on the other side of a reasonable debate ordinarily would do greater harm to the rule of law than leaving them settled.

The calculus is different for past decisions of the “hasty and crude” variety. If a past decision ignores or flatly disregards sound, generally accepted legal principles, or relies only on bald, unreasoned assertions, or some combination of the above, the inference that such decisions are proper exercises of the judicial power grounded in law is much weaker.20 Following those decisions

20 I would also tend to include in this category past decisions that “uncritically import” holdings of federal courts into state law. Buckner-Webb v. State, 314 Ga. 823, 834 (1) (878 SE2d 481) (2022) (Pinson, J., concurring) (quoting Elliott, 305 Ga. at 188 (II) (C)). See also id. (“When we rely on such

51

when they are probably wrong poses risks of undermining the rule of law similar to the risk posed by overruling—that is, it suggests that courts are relying on arbitrary discretion or personal preferences rather than following the law that the people, or our elected representatives, enacted. Our Court has not hesitated to overrule such decisions. See, e.g., Jackson, 287 Ga. at 653 (3) (overruling State v. Crane, 247 Ga. 779 (279 SE2d 695) (1981), and noting that the “one-and-a-half page opinion . . . did not consider the customary legal meaning of ‘cause’ or look to our then-existing case law interpreting that term,” but instead “baldly asserted” that it could choose one of two interpretations and picked the defendant- friendly one “[b]ecause a criminal statute was being interpreted”); Gilliam v. State, 312 Ga. 60, 63 (860 SE2d 543) (2021) (overruling decision that took jurisdiction over certain appeals for “judicial economy,” “ignor[ing] the constitutional parameters of its jurisdiction without any significant analysis”); State v. Hudson, 293

federal decisions without making sure the relevant text and context match up, we risk giving an ‘interpretation’ of Georgia law that is arbitrary, wrong, or both.”).

52

Ga. 656, 661-62 (748 SE2d 910) (2013) (overruling decision that “contain[ed] no analysis supporting its adoption of the count-by- count approach but instead adopt[ed] that approach as though there were no other alternative” (emphasis in original)).

This distinction just discussed may not be the only thing that matters to the question whether to overrule a past decision, but in my view, it serves a kind of gatekeeping function in any stare decisis analysis. If the past decision in question is unreasoned, or if it disregards the basic legal principles that courts use to do law, the argument for overruling is easier to make. See, e.g., Crayton v. State, 298 Ga. 792, 803 (784 SE2d 343) (2016) (Blackwell, J., dissenting) (disapproving of a holding reached “without any discussion or analysis whatsoever” and explaining that “[w]e ought not follow unreasoned precedent without reason”). If a past decision is “not law,” Doe, 23 Ga. at 86, treat it accordingly. On the other hand, if the past decision in question is the product of the careful and deliberate application of sound and accepted legal principles, it seems to me that the burden on any would-be overrulers is to show

53

something pretty extraordinary to justify the serious harm to the rule of law that comes from overruling that kind of decision.

2. That brings us to this case. The majority holds that a person’s right against compelled self-incrimination under the Georgia Constitution, Ga. Const. of 1983, Art. I, Sec. I, Par. XVI (“Paragraph XVI”), prevents the State from using that person’s refusal to perform preliminary breath tests and certain field sobriety tests against her at trial to suggest an adverse inference of guilt. That holding follows directly from this Court’s recent decisions in Olevik, 302 Ga. 228, Elliott, 305 Ga. 179, and Awad, 313 Ga. 99. In Olevik and Elliott, we concluded, and then reaffirmed, that the right against compelled self-incrimination prevents the State from forcing people to take affirmative acts that inherently generate incriminating evidence, and we applied that holding in Awad.

There is perhaps room for debate about whether these past decisions are correct as an original matter. (This is apparent from the existence of a dissent in this case that engages in that debate.) But there can be no serious dispute that these decisions—and here

54

I focus on Olevik and Elliott—are very much the “deliberate” kind I’ve just described. In Olevik, a unanimous Court considered and explained in detail the set of objective and well-established legal principles that formed the framework for analyzing the question whether Paragraph XVI applied to a chemical breath test. See Olevik, 302 Ga. at 235-39 (2) (c) (i). The Court then carefully applied those principles and, after canvassing more than 100 years of decisional law and constitutional language to do so, concluded that Paragraph XVI prevents the State from forcing people to take affirmative acts that inherently generate incriminating evidence. Id. at 239-41 (2) (c) (ii). Two years later, in Elliott, a unanimous Court did all of that again in an even more expansive analysis after the State asked the Court to overrule Olevik, and then reaffirmed its holding in Olevik. Elliott, 305 Ga. at 181-209 (II)-(III). The Court then addressed, in similarly exhaustive fashion, the separate question whether Paragraph XVI prevents the government from using a person’s refusal to perform protected affirmative acts against her at trial to suggest an adverse inference of guilt, and

55

concluded that it did. Id. at 209-21 (IV) (A)-(D). Whatever one’s views about how to answer the questions these two decisions addressed as an original matter, it is not possible to read them and come away thinking that how they addressed and resolved those questions is anything other than consistent with the rule of law.

Given the deliberate nature of these decisions, anyone who seeks to overrule them has to marshal much more than mere disagreement with their outcome—to me, they need to show in some way that following them would cause even more serious damage to the rule of law than overruling them would. The dissent has not nearly made that case. Although the dissent applies the familiar four-factor analysis for assessing whether to apply stare decisis,21 its

21 For what it’s worth, I am not sure that rote application of these four

factors is ever all that helpful to deciding whether stare decisis applies to a given past decision. Courts apply these factors because we understand that overruling a past decision must be supported by something more than just an argument that it is wrong. See, e.g., Nalls v. State, 304 Ga. 168, 179 (3) (b) (815 SE2d 38) (2018) (considering whether stare decisis factors “counsel[ed] us not to” overrule prior case law even though we concluded it was “incorrectly decided”). But these factors do little to address the big reason, in my view, for that understanding: that overruling past decisions risks undermining the rule of law. The “reliance” and “workability” factors largely address practical concerns in theory, and they are inherently malleable in practice, allowing

56

arguments reduce to mere disagreements with how those decisions should have applied the relevant legal principles, hypotheticals that might pose close questions in the future, a couple of past cases that are arguably inconsistent with Olevik and Elliott, and the policy concern that the General Assembly has been “stripped . . . of its authority to protect the public from dangerous drivers.”22 This would be a perfectly fine dissent from Olevik. But it does not come close to justifying the harm to the rule of law of overruling that unanimous,

courts to instead raise any number of policy concerns that are more appropriate for the legislature to address. The “soundness of reasoning” factor might roughly address the rule-of-law concern if the analysis focused more on the process and legal framework that undergirds the past decision, but courts more often use it to restate mere disagreements with how a past court applied accepted legal principles. And a precedent’s age never actually seems to make a difference to the analysis. At the least, I agree with my colleagues who have recognized that the analysis of whether stare decisis applies to a given decision is not limited to this list of four factors. See Cook, 313 Ga. at 509-10 (Peterson, J., dissenting).

22 The dissent says that the Court has “stripped” the General Assembly

of authority to protect the public from dangerous drivers. The other way of stating it is that the Court recognized after careful deliberation that the people of Georgia ratified a constitutional right that limits the State’s authority to compel citizens to incriminate themselves through affirmative acts. And of course, if enough of the people share the policy view that the State should be allowed to do this in the context of drunk driving, they can amend the Georgia Constitution to allow the General Assembly to do that. See Elliott, 305 Ga. at 225 (Boggs, J., concurring) (explaining that, if the General Assembly and the people are unhappy with the meaning of a constitutional provision, they are free to amend the Constitution).

57

carefully reasoned decision just five years later, especially when a unanimous court has since reaffirmed it, and nothing material to the legal question has changed—only the Court’s personnel.

For my part, I am satisfied that stare decisis applies to Olevik and Elliott. I have a lingering question or two about certain aspects of the reasoning of those decisions, but their holdings are plainly grounded in careful—indeed, exhaustive—application of sound and generally accepted legal principles, and I see nothing in the arguments of the dissent or the parties that suggests following those decisions has caused or is likely to cause any substantial harm to the rule of law.23 For that reason, I concur in the majority’s faithful application of those decisions here.

23 To avoid being cryptic: my lingering questions have to do with the

Court’s application of the prior-construction canon in those cases. That canon says that when language is enacted that had previously received an authoritative construction by a jurisdiction’s court of last resort, that language is understood according to the prior construction. See Olevik, 302 Ga. at 237 (2) (c) (i) (citing Scalia & Garner, Reading Law: The Interpretation of Legal Texts 322-26 (West 2012)). The Court in Olevik and Elliott rightly explained that we demand a “consistent and definitive construction” of the old constitutional language to trigger the presumption that the reenacted language carried forward that construction. Elliott, 305 Ga. at 184 (II) (B). In applying that rule, the Court marshaled a lot of decisions that together show

58

I am authorized to state that Justice Warren joins in this concurrence.

a consistent and definitive conclusion that Paragraph XVI protected affirmative acts. Id. at 202-05 (III) (C) (i). But as far as I can tell, these decisions did precious little “construction” of the actual language of the prior versions of Paragraph XVI—that is, they did not seriously engage the question of what the actual language of the clause meant to the public at the time it was enacted. Because the rationale behind the prior-construction canon depends on finding a prior construction of the language that we presume the people or legislature were aware of and carried forward, I am not sure how much weight decisions like these deserve in this analysis, or just how any such absence of meaningful construction bears on the related question whether the presumption that the past construction is carried forward is rebutted. Id. at 186 (II) (B) n.6 (declining to “articulate precisely when such a presumption may be rebutted”). But I also can’t say that the Court’s deliberate resolution of these and other difficult questions of constitutional interpretation in Olevik or Elliott was unreasonable (much less unreasoned), or that following it risks any kind of lasting or significant harm to the rule of law.

59

COLVIN, Justice, concurring in part and dissenting in part.

Article I, Section I, Paragraph XVI of the Georgia Constitution of 1983 (“Paragraph XVI”) provides that “[n]o person shall be compelled to give testimony tending in any manner to be self- incriminating.” By its plain terms, this provision protects only the right against compelled self-incriminating testimony. Yet we have disregarded this express textual limitation, construing the constitutional right in a manner inconsistent with the constitutional text and extending the right to all compelled self-incriminating acts. I have previously expressed “grave concerns” about our construction of Paragraph XVI, Awad v. State, 313 Ga. 99, 107 (868 SE2d 219) (2022) (Colvin, J., concurring), and the State now squarely asks us to reconsider our expansive reading of that provision. Because I believe this Court’s interpretation of Paragraph XVI and its predecessors contradicts the constitutional text and lacks any persuasive justification, I would overrule our precedent and clarify that the scope of the constitutional right is limited to “testimony.” Further, because submitting to, or refusing to submit to, a chemical

60

or field sobriety test does not require a defendant “to give testimony,” I do not believe that Paragraph XVI prohibits the State from admitting into evidence the results of, or refusal to submit to, a state-administered chemical or field sobriety test. Accordingly, while I concur with Divisions 1 and 3 of the majority opinion, I dissent with respect to Division 2.

In explaining the interpretive principles relevant to construing this State’s Constitution, we have emphasized the need to ascertain the “original public meaning” of a constitutional provision, that is, “the meaning the people understood a provision to have at the time they enacted it.” Olevik v. State, 302 Ga. 228, 235 (2) (c) (i) (806 SE2d 505) (2017). As we have explained, this task requires an objective inquiry that “consider[s] the plain and ordinary meaning of the text, viewing it in the context in which it appears and reading the text in its most natural and reasonable manner.” Id. at 236 (2) (c) (i). Yet, for most of our history, we have given little consideration to the plain and ordinary meaning of the text when construing the constitutional right against self-incrimination. Indeed, in our recent

61

precedent, we have relegated to a footnote Drake v. State, 75 Ga. 413 (1885), the first case in which we expressly construed the text of the self-incrimination provision, and dismissed as irrelevant the question of whether we got it right when we offered a contrary construction in other decisions following the 1877 Constitution’s ratification. See Elliott v. State, 305 Ga. 179, 203 (III) (C) (i) n.16 (824 SE2d 265) (2019) (disregarding Drake); see id. at 209 (III) (C) (ii) (“[W]e do not determine conclusively that [Day v. State, 63 Ga. 667 (1879)] was correctly decided[.]”). See also Olevik, 302 Ga. at 241 (2) (c) (ii) (assuming arguendo that our early precedent misread the constitutional text).

If we give any weight at all to the specific language used in the Constitution, it becomes impossible to conclude that the constitutional right against self-incrimination extends to all incriminating acts. The right against self-incrimination first appeared in Georgia’s 1877 Constitution. At the time, the provision read: “No person shall be compelled to give testimony tending in any manner to criminate himself.” Ga. Const. of 1877, Art. I, Sec. I, Par.

62

VI (emphasis supplied). The language of this provision was incorporated into the 1945 and 1976 Constitutions without change and was not materially altered when, in 1983, our current Constitution replaced the outdated phrase “to criminate himself” with the more modern phrase “to be self-incriminating.” Ga. Const. of 1945, Art. I, Sec. I, Par. VI; Ga. Const. of 1976, Art. I, Sec. I, Par. XIII; Ga. Const. of 1983, Art. I, Sec. I, Par. XVI; see Olevik, 302 Ga. at 239 (2) (c) (ii) n.5 (noting that “criminate” is “merely an archaic variant of ‘incriminate’”). The constitutional text recognizing a right against self-incrimination has therefore always limited the scope of that right to “testimony.”

The meaning of “to give testimony” has not significantly changed since the phrase first appeared in our 1877 Constitution. See Olevik, 302 Ga. at 239 (2) (c) (ii) n.6 (“There is no indication that ‘testimony’ had a substantially broader definition in 1877.”). At the time, Noah Webster defined “testimony” as “[a] solemn declaration made to establish some fact,” as exemplified by “the evidence of a witness given under oath.” Noah Webster, A Dictionary of the

63

English Language 434 (1878). John Guerard invoked this ordinary sense of the word when, at the 1877 constitutional convention, he introduced the text of the constitutional right against compelled self- incrimination and explained why it was a necessary addition. Absent this constitutional guarantee, Guerard explained, “a man may be subjected to an inquisition, and made to testify against himself.” Samuel W. Small, A Stenographic Report of the Proceedings of the Constitutional Convention Held in Atlanta, Georgia, 1877 94 (Constitution Publishing Company 1877) (emphasis supplied); see Noah Webster, A Dictionary of the English Language 226 (1878) (defining “inquisition” as (1) “[i]nquiry; investigation,” (2) “[j]udicial inquiry,” or (3) “[a] tribunal for examining and punishing heretics”). See also Olevik, 302 Ga. at 238 (2) (c) (i) (“[C]onsidering what the framers of our Constitution understood the words they selected to mean can be a useful data point in determining what the words meant to the public at large.”). I am aware of no evidence suggesting that the public in 1877 would have understood the term “testimony,” as used in the constitutional

64

self-incrimination provision, to have some technical or particular idiosyncratic meaning not captured in dictionaries of the time. See Olevik, 302 Ga. at 239 (2) (c) (ii) n.6 (relying on the 1878 Noah Webster dictionary definition for the original public meaning of the word “testimony”).24 This Court has often referred to Day v. State, 63 Ga. 667 (1879)

as the seminal case construing the constitutional right against self- incrimination. See, e.g., Olevik, 302 Ga. at 239 (2) (c) (ii). Day, however, merely suggested that the constitutional right and an associated common-law right provided related protections:

By the constitution of this state “no person shall be compelled to give testimony tending in any manner to criminate himself.” Nor can one, by force, compel another, against his consent, to put his foot in a shoe-track for the purpose of using it as evidence against him on the criminal side of the court, the more especially when the person using such force has no lawful warrant or authority for doing so.

24 Although the 1877 Constitution no longer governs, the ordinary meaning of “testimony” has not significantly changed in the intervening years. In 1982, when the people of Georgia ratified the current Constitution, “testimony” was defined as “a solemn declaration usu[ally] made orally by a witness under oath in response to interrogation by a lawyer or authorized public official” or “firsthand authentication of a fact.” Webster’s Ninth New Collegiate Dictionary 1219 (1985).

65

Day, 63 Ga. at 669 (2) (emphasis supplied). Neither this statement nor Day’s headnote, which stated that “[a] defendant cannot be compelled to criminate himself by acts or words,” purported to construe the constitutional provision. Id. at 667 (2). Moreover, Day’s holding appears to rely in part on the requirement that the State obtain a “lawful warrant” before obtaining evidence from a defendant against his will—a requirement that has no basis in the constitutional self-incrimination provision. Id. at 669 (2).

Given that “to give testimony” had a clear meaning in 1877, it is unsurprising that, when first called upon to construe the constitutional text in Drake, we concluded that the self- incrimination provision meant exactly what it said. The constitutional provision declaring that “no person shall be compelled to give testimony tending in any manner to criminate himself,” we explained, “means that, when a person is sworn as a witness in a case, he shall not be compelled to testify to facts that may tend

66

to criminate him.”25 Drake, 75 Ga. at 414-415 (punctuation omitted) (quoting Ga. Const. of 1877, Art. I, Sec. I, Par. VI) (holding that “[i]t would be a forced construction” of the constitutional provision “to hold that clothing or any other article taken from a person accused of crime could not be given in evidence or exhibited to the jury” (emphasis supplied)).26 While there may be factual circumstances in which reasonable minds might differ as to whether the evidence at issue constitutes “testimony,” the text of the constitutional provision is not consistent with an interpretation that the constitutional right protects against any and all compelled self-incriminating “acts.” In concluding otherwise, this Court has expressly disregarded the specific

25 Because Day had not construed the constitutional provision, Drake had

no reason to cite, discuss, or distinguish Day.

26 Nine years later, in Rusher v. State, we commented again that it was

“manifest” from “the letter” of the constitutional self-incrimination provision that it concerned “the giving of testimony by the accused” rather than “evidence of facts, acts and declarations known to and detailed by other witnesses.” 94 Ga. 363, 366 (1) (21 SE 593) (1894) (emphasis supplied). We also rejected as “unsound” an argument “that the spirit of the constitutional provision extends to anything which a person under accusation, or afterwards accused, is coerced to do or say out of court before trial, or in court during the trial.” Id. at 366- 367.

67

language of the self-incrimination provision, relying instead on two non-textual inferences: first, because the constitutional right against self-incrimination derived from the common-law right against self-incrimination, the constitutional right is identical to the common-law right; and second, even if our early precedent badly misinterpreted the constitutional provision, our incorrect construction became the “original public meaning” of the provision when it was carried forward into a later constitution without material change. In my view, neither of these inferences is sound.

Calhoun v. State was the first decision of this Court to conflate the constitutional and common-law rights based on faulty logic. In Calhoun, we explained that the common-law privilege had “derived” from, and had been “uniformly construed” as coextensive with, the common-law maxim “that no man is bound to accuse himself of any crime or to furnish any evidence to convict himself of any crime.” 144 Ga. 679, 680 (87 SE 893) (1916). We further explained that this maxim was the “prototype” of “the constitutional mandate that ‘[n]o person shall be compelled to give testimony tending in any manner

68

to criminate himself.’” Id. Then, based on the historical fact that the common-law right was a precursor of the constitutional right, we leapt to the conclusion that

[t]he constitutional guaranty protects one from being compelled to furnish evidence against himself, either in the form of oral confessions or incriminating admissions of an involuntary character, or of doing an act against his will which is incriminating in its nature.

Id. at 680-681 (citing Day, 63 Ga. 667). In other words, Calhoun concluded that the constitutional right was “as broad as that afforded by the common-law principle from which it is derived” simply because the two rights were historically associated. Id. at 680. This does not follow. Calhoun did not analyze the constitutional language to determine whether the text could be fairly interpreted as encompassing the full scope of the common-law right not to be compelled to furnish evidence against oneself. Nor did it attempt to explain why Drake had erred in construing the constitutional text as limited to compelled self-incriminating

69

testimony.27 Although this Court has implicitly endorsed Calhoun’s reasoning, see Olevik, 302 Ga. at 239-240 (2) (c) (ii), we have never offered a robust defense of the proposition that the constitutional right against self-incrimination is identical to the common-law right

27 Calhoun is one of several cases in which this Court has ignored Drake’s

construction of the self-incrimination provision as irrelevant to its holding. See, e.g., Elliott, 305 Ga. at 203 (III) (C) (i) n.16 (stating that Drake’s construction of the constitutional provision, “though sounding like a holding, was not the actual holding of the case, because the Drake Court held that taking clothing from a defendant and submitting that clothing to the jury would not violate the defendant’s constitutional right against compelled self- incrimination”); Calhoun, 144 Ga. at 681 (“While the headnote in the case of [Drake], supra, restricts the application of the constitutional privilege to persons sworn as witnesses, an examination of the facts of the case will show that the actual ruling was that the constitutional privilege does not prevent the introduction in evidence or the exhibition to the jury of clothing or any other article taken from a person accused of crime, where they tend to show his guilt.”); Evans v. State, 106 Ga. 519, 521 (32 SE 659) (1899) (“While the headnote in the case of [Drake], 75 Ga. 413, restricts the application of the constitutional provision above quoted to persons sworn as witnesses in a case, an examination of the facts appearing of record in that case will show that it is really not in conflict with the Day case, or the ruling made in the present case.”). Drake’s construction of the constitutional provision, however, was essential to its holding, as it was the only reason the Court gave for reaching its ultimate conclusion. See South Ga. Med. Center v. Washington, 269 Ga. 366, 367 (1) (497 SE2d 793) (1998) (“An adjudication on any point within the issues presented by the case cannot be considered a dictum, and this rule applies as to all pertinent questions, although only incidentally involved, which are presented and decided in the regular course of the consideration of the case, and lead up to the final conclusion, and to any statement in the opinion as to a matter on which the decision is predicated.” (citation andpunctuation omitted)).

70

from which it derived. I do not dispute that the constitutional right’s historical predecessor was the common-law right or that the common law broadly recognized a right not to be compelled to furnish evidence against oneself by words or acts. See Marshall v. Riley, 7 Ga. 367, 370-371 (3) (1849) (describing “[t]he maxim of the Common Law . . . that no man is bound to accuse himself of any crime, or to furnish any evidence to convict himself of any crime” (emphasis in original)). But Georgia’s Constitution refers to the right against being compelled “to give testimony,” not the right against being compelled “to furnish evidence.” Concluding that the specific language incorporated into this State’s Constitution has no impact on the scope of the resulting constitutional right conflicts with fundamental principles of constitutional interpretation. Where, as here, the language of the constitutional text differs from the more expansive language used at common law, we should not equate the common-law right with the right protected by our

71

Constitution.28 As an alternative basis for construing the constitutional right against self-incrimination without regard to the Constitution’s text, this Court has relied on a version of the “prior-construction canon,” which generally provides that, “[i]f a statute uses words or phrases that have already received authoritative construction by the jurisdiction’s court of last resort, . . . they are to be understood according to that construction.” Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 322 (2012)

28 The only “textual” justification that this Court has offered for incorporating the entirety of the common-law self-incrimination right into the Constitution does not withstand scrutiny. We have explained that, “where the right enshrined in the constitution was one found at common law, that constitutional right is understood with reference to the common law, absent some clear textual indication to the contrary.” Elliott, 305 Ga. at 212 (IV) (B) (emphasis supplied). This statement purports to leave open the possibility that the specific words used in the constitutional text might have some impact on the scope of the resulting constitutional right. But we have foreclosed that possibility. As we have explained, “textual differences” in how different states have constitutionalized the common-law right against self-incrimination (that is, whether a self-incrimination provision refers to furnishing evidence, giving testimony, or being a witness) are “understood as not reflecting a difference in meaning because they all refer to the same common law.” Id. at 196 (III) (B). In other words, we simply assume that any reference to a common-law right in the Constitution naturally incorporates the entire common-law right, regardless of whether one can in fact fairly read the constitutional text as incorporating the entire common-law right.

72

(hereinafter “Reading Law”). Adapting that principle for purposes of constitutional interpretation, we have explained that there is a presumption that, when a constitutional provision is incorporated into a new or amended constitution without material change, a “consistent and definitive construction” of the constitutional provision — even if flat out wrong — is carried forward as the meaning of the new or amended constitution. See Elliott, 305 Ga. at 184 (II) (B) (“Given th[e] consistent and definitive construction [of the constitutional self-incrimination provision, Olevik] presumed that construction was carried forward into the 1983 Constitution.”). See also Olevik, 302 Ga. at 241 (2) (c) (ii) (“[E]ven if we were wrong in Day and Calhoun . . . the subsequent ratifications of new constitutions with the same language are strongly presumed to have carried forward the interpretation of that language provided by Day and Calhoun.”).

We first applied a version of this presumption to the self-

incrimination provision in Aldrich v. State, 220 Ga. 132 (137 SE2d 463) (1964). There, we remarked that, “[f]ortunately” for us, we did

73

not have to wrestle with the constitutional text to determine “whether or not ‘testimony’ as found in the Constitution embraces all kinds of evidence.” Id. at 134. Making no mention of Drake’s holding that “testimony” meant “testimony,” we noted that “this court has many times decided that question by holding that the word ‘testimony’ means all types of evidence.” Id. at 134. Then, we simply applied a presumption “that the framers of [a] Constitution intend[ ] for [an identical constitutional provision carried forward into a new constitution] to have the meaning theretofore given it by construction.” Id. at 135. With that, Calhoun’s expansive construction of the constitutional right against self-incrimination was incorporated into the 1945 Constitution as the self- incrimination provision’s definitive interpretation.29 Later, in Olevik, we followed similar logic, presuming that Calhoun’s

29 Neither Day nor Calhoun purported to construe the specific word “testimony” (as opposed to the self-incrimination provision as a whole), and the only decision of this Court to do so was Drake (which held that the word “testimony” in fact meant “testimony”). Nevertheless, Aldrich concluded that Day, Calhoun, “and many more decisions of this court had construed the word ‘testimony’ to embrace any evidence when the identical clause containing this word was written into the 1945 Constitution.” Aldrich, 220 Ga. at 134-135.

74

interpretation of the constitutional right against compelled self- incrimination was carried forward as the meaning of the 1983 Constitution’s self-incrimination provision. See Olevik, 302 Ga. at 241 (2) (c) (ii).

To be sure, applying such a presumption has some pragmatic benefits, making our jobs easier and maintaining consistency in our rulings. See Elliott, 305 Ga. at 186 (II) (B) (“The presumption created by a consistent and definitive construction reflects the value of consistency in the interpretation of legal language.” (citation and punctuation omitted)). But even those who advocate for the presumption as an interpretive tool admit that it comes at the potential “cost” of permanently enshrining into law a high court’s prior incorrect construction, and that such a consequence should be “avoided when the application of other sound rules of interpretation overcomes this canon.” Reading Law at 324.30

30It is unclear whether this prior-construction presumption is a methodologically sound tool for interpreting constitutional text within the framework this Court has set out for proper constitutional interpretation. We have said that the Constitution must “be construed in the sense in which it

75

This Court’s overriding reliance on the prior-construction canon in the self-incrimination context is at odds with the fundamental principle that “[n]o canon of interpretation is absolute” and “[e]ach may be overcome by the strength of differing principles

was understood by the makers of it at the time when they made it,” and that “the people” who ratified the Constitution “are the ‘makers’ of the Georgia Constitution.” Olevik, 302 Ga. at 235-236, 238 (2) (c) (i) (citations, punctuation and emphasis omitted). In other words, the meaning of a constitutional provision is the meaning that the voters who ratified the Constitution would ascribe to the provision. See id. at 238 (2) (c) (i) (noting that constitutional interpretation seeks to “determin[e] what the words meant to the public at large,” that is, the “citizens who voted on its ratification”). See also Clarke v. Johnson, 199 Ga. 163, 164 (33 SE2d 425) (1945) (“Constitutions are the result of popular will, and their words are to be understood ordinarily in the sense they convey to the popular mind.” (citation and punctuation omitted)). But when the State in Elliott pointed out that there was no evidence that “the public understood” our prior construction of the self-incrimination provision “when ratifying the 1983 Constitution,” Elliott, 305 Ga. at 206-207 (III) (C) (ii), this Court had to redefine the nature of the interpretive inquiry to avoid the logical implications of the State’s argument. Although we had previously said that the understanding of the “citizens who voted on [the Constitution’s] ratification” governed the meaning of the Constitution, Olevik, 302 Ga. at 238 (2) (c) (i), we criticized the State for focusing on how “citizen[s] understood the particular meaning of a constitutional provision,” Elliott, 305 Ga. at 207 (III) (C) (ii). Instead, we said, “it is the understanding of the text by reasonable people familiar with its legal context that is important.” Elliott, 305 Ga. at 207 (III) (C) (ii) (citation and punctuation omitted; emphasis supplied). Clearly, however, whether it is reasonable to assume that a reader of constitutional text will understand a provision in accordance with this Court’s construction of a materially identical provision from a prior constitution largely depends upon whose perspective we adopt — voters at large or well-informed lawyers.

76

that point in other directions.” Reading Law at 59.31 When it comes to Paragraph XVI, the prior-construction canon is the only interpretive principle that favors this Court’s conclusion that the provision applies to all compelled self-incriminating acts. Every other applicable principle of textual interpretation points strongly in the opposite direction, including the supremacy-of-text principle, id. at 56 (“The words of a governing text are of paramount concern, and what they convey in their context, is what the text means.”), the ordinary-meaning canon, id. at 69 (“Words are to be understood in their ordinary, everyday meanings — unless the context indicates that they bear a technical sense.”), and the negative-implication canon, id. at 107 (“The expression of one thing implicates the exclusion of others (expressio unius est exclusio alterius).”).

As relevant here, the key phrase in Paragraph XVI is “to give testimony,” and nothing about the textual context in which that

31 Although we have insisted that the prior-construction presumption is

“rebuttable,” we have declined to specify how it might be rebutted. See Elliott, 305 Ga. at 186 (II) (B) & n.6 (“[T]his is not a case that calls us to articulate precisely when such a presumption may be rebutted.”).

77

phrase appears — “[n]o person shall be compelled to give testimony tending in any manner to be self-incriminating” — suggests that the right protected by Paragraph XVI applies to something other than “testimony.” Further, the word “testimony” has a commonly understood meaning that has changed remarkably little since it first appeared in the Georgia Constitution, and no one contends that the Georgians who ratified Paragraph XVI of the 1983 Constitution would have understood the term to carry some technical or archaic sense. See Reading Law at 69 (“Interpreters should not be required to divine arcane nuances or to discover hidden meanings.”). Finally, Paragraph XVI’s use of the phrase “to give testimony” — to the exclusion of other, broader phrases frequently used in connection with the common-law right against self-incrimination — gives rise to a negative implication that the scope of Paragraph XVI’s protections is narrower than the protections afforded by the common law. In sum, even assuming that the prior-construction canon should be afforded some weight in interpreting Paragraph XVI, other sound principles of interpretation overwhelmingly favor an

78

interpretation of Paragraph XVI that affords protection only to compelled self-incriminating “testimony.” An interpretation that extends the scope of Paragraph XVI to all compelled self- incriminating “acts” is simply incompatible with the constitutional text.

As this Court has recognized, stare decisis is neither “a straightjacket,” State v. Jackson, 287 Ga. 646, 647 (697 SE2d 757) (2010), nor “an inexorable command,” Cook v. State, 313 Ga. 471, 485 (3) (a) (870 SE2d 758) (2022) (citation and punctuation omitted). In determining whether to overrule a prior erroneous ruling, we have considered a variety of factors, including “the age of precedent, the reliance interests at stake, the workability of the decision, and, most importantly, the soundness of its reasoning.” Gilliam v. State, 312 Ga. 60, 62 (860 SE2d 543) (2021) (citation and punctuation omitted). Further, “it is well settled that stare decisis applies with the least force to constitutional precedents” because “it is much harder for the democratic process to correct or alter our interpretation of the Constitution than our interpretation of a

79

statute or regulation.” Id. (citations and punctuation omitted). As a result, “[t]he more wrong a prior precedent got the Constitution, the less room there is for the other factors to preserve it.” Id. at 62- 63 (citation and punctuation omitted).

Here, age is the only stare decisis factor that weighs in favor of retaining our precedent holding that the constitutional right against compelled self-incrimination applies not only to “testimony” but also to “acts.” Whether one measures from Calhoun or Day, our erroneous precedent stretches back more than 100 years. That is undeniably old precedent. But it is not ancient, and this Court is no stranger to overruling old precedent that is demonstrably wrong. See, e.g., Frett v. State Farm Employee Workers’ Compensation, 309 Ga. 44, 60 (3) (c) (844 SE2d 749) (2020) (overruling 85-year-old statutory precedent to which stare decisis applied with more force).32 The remaining stare decisis factors all weigh in favor of

32 I assume here that our pre-1983 precedents are relevant to our stare

decisis analysis, even though they offered interpretations of self-incrimination provisions in prior constitutions that are no longer in force. Olevik, our first decision to definitively construe the 1983 Constitution’s self-incrimination provision, issued only five years ago.

80

overruling our self-incrimination precedent. Our erroneous decisions “created no reliance interest of the sort normally given weight in stare decisis analysis.” Gilliam, 312 Ga. at 63; see Olevik, 302 Ga. at 245 (2) (c) (iv) (“Substantial reliance interests are an important consideration for precedents involving contract and property rights, where parties may have acted in conformance with existing legal rules in order to conduct transactions.” (citation and punctuation omitted)). Further, while overruling our self- incrimination precedent might negatively impact some people with pending criminal charges against them, most of those impacts would not implicate reliance interests: those who were compelled to perform a self-incriminating act obviously did not rely on our holdings that they could not be compelled to perform such acts; and there are presumably few, if any, people currently facing criminal charges who refused to perform a self-incriminating act because our holdings led them to believe that their refusal could not be admitted into evidence against them.

The workability factor also weighs in favor of overruling our

81

erroneous self-incrimination precedent. Our decisions in this area demonstrate that we have failed to formulate a non-arbitrary standard for when a defendant was compelled to engage in a self- incriminating act. Examples abound. We have characterized Day as exemplifying an affirmative act because the defendant “was compelled to place his foot in certain footprints located near the crime scene.” Olevik, 302 Ga. at 241 (2) (c) (iii). Yet, placing his foot in the footprint could not have required any more than a de minimis act on the part of the defendant, given that an agent of the State “took hold of his foot and put it in the track.” Day, 63 Ga. at 669 (2). At most, what was required of the defendant in Day was that he maintain his balance while the State “forcibly” moved his body. Id. at 667 (2).

We have also said that requiring a defendant to “stand up at trial” so a witness could look at his amputated leg requires an affirmative act, Olevik, 302 Ga. at 241 (2) (c) (iii) (citing Blackwell v. State, 67 Ga. 76, 78-79 (1881)), but that requiring a defendant to “strip to the waist” so police could photograph his tattoos did not

82

require an affirmative act, id. at 242 (2) (c) (iii) (citing Ingram v. State, 253 Ga. 622, 634 (7) (323 SE2d 801) (1984)). This makes little sense, given that the act of stripping is more involved than the act of standing, and in both cases the evidence obtained was merely a visual inspection of the defendant, rather than something the State or the defendant removed from the defendant’s body.

In addition, we have described taking dental impressions as a method of evidence collection that only requires a defendant “to be present” so evidence can be “taken from [his] body.” Olevik, 302 Ga. at 242 (2) (c) (iii) (citing State v. Thornton, 253 Ga. 524, 525 (2) (322 SE2d 711) (1984)). What this ignores, however, is that taking dental impressions requires significant cooperation on the part of a defendant, who could easily prevent the State from obtaining a usable impression by refusing to open his mouth or moving his jaw during the procedure.

Further, we have held that a defendant performs an “act”

under Paragraph XVI if, “at the time and in the manner directed by the State,” he “urinate[s] into a collection container to generate a

83

sample for chemical testing.” Awad, 313 Ga. at 103 (3).33 This is so even though a defendant held in a jail cell against his will presumably has not performed a compelled self-incriminating “act” if he “chooses” to use whatever restroom facilities are provided, thereby generating a urine sample for chemical testing without being directed to do so by the State.

Lurking in the record here is yet another self-incrimination issue that will require this Court to engage in arbitrary line drawing. Specifically, during the traffic stop here, State Trooper Levi Perry required Mia Ammons to produce her driver’s license, which revealed that she had not timely updated her address information after moving. As a result, Ammons was charged with violating OCGA § 40-5-33. While Ammons has not argued that Paragraph XVI prevents a law enforcement officer from requiring a driver to produce a driver’s license during a traffic stop, we will

33 I authored the Awad majority opinion, “faithfully appl[ying] this Court’s recent precedent interpreting Georgia’s constitutional right against compelled self-incrimination because the State argued only that its position was consistent with that precedent and not that the Court should reconsider it.” Awad, 313 Ga. at 106-107 (Colvin, J., concurring).

84

inevitably have to confront such an argument if this Court stays the course with its self-incrimination case law. Perhaps this Court will hold that, because a driver does not create a driver’s license but merely provides the license to an officer during a traffic stop, giving a license to an officer is not “an act that itself generates incriminating evidence.” Olevik, 302 Ga. at 243 (2) (c) (iii) (emphasis supplied). But if that is the case, another rift in our case law will develop, as we have said that a person engages in an affirmative act when, at the direction of a law enforcement officer, he reaches into his pocket to produce a pistol. See Elliott, 305 Ga. at 203 (III) (C) (i) (citing Evans, 106 Ga. at 521).

As these examples demonstrate, the rule established by our precedent — that a defendant’s Paragraph XVI right is violated if he is compelled “to perform an act that itself generates incriminating evidence,” Olevik, 302 Ga. at 243 (2) (c) (iii) — cannot be consistently and non-arbitrarily administered. The workability factor therefore counsels against retaining our precedent.

The final stare decisis factor — the soundness of our

85

precedent’s reasoning — strongly favors overruling this Court’s erroneous interpretations of Paragraph XVI. This is the “most important factor” and a “critical” one when it comes to whether the stare decisis analysis favors retaining a prior decision’s interpretation of the Constitution. Olevik, 302 Ga. at 245 (2) (c) (iv). In concluding that the scope of the constitutional right against self- incrimination extends to all compelled self-incriminating acts, our recent precedent relied on older cases, such as Calhoun, that construed the constitutional text without performing the necessary textual analysis. Unlike Drake, which reasonably construed the self-incrimination provision as limited to “testimony” based on the constitutional text’s plain and ordinary meaning, Calhoun did not purport to analyze the text at all. Rather, Calhoun fallaciously reasoned that the constitutional and common-law rights against self-incrimination were identical because they were historically associated. This sort of atextual analysis would not pass muster today, nor should it. Yet, it is Calhoun’s incorrect construction of the constitutional right against self-incrimination, rather than

86

Drake’s correct one, that this Court continues to endorse.

Charitably reading our recent precedent, we have implicitly admitted that Calhoun’s construction of the self-incrimination provision was incorrect. See Olevik, 302 Ga. at 235 (2) (c) (noting that we might interpret the constitutional provision differently “[i]f we were construing Paragraph XVI in the first instance”). Nevertheless, we have relied on a version of the prior-construction canon to retain our incorrect interpretation of the self-incrimination provision. As discussed above, our reliance on the prior-construction canon to the exclusion of competing canons of construction conflicts with fundamental principles of constitutional interpretation. Every relevant interpretive principle other than the prior-construction canon strongly suggests that Calhoun’s interpretation of the self- incrimination provision was wrong. Accordingly, our precedent adopting Calhoun’s interpretation as the definitive construction of Paragraph XVI of the 1983 Constitution is unsound.

As illustrated by this Court’s recent opinions in DUI cases, our misinterpretation of the constitutional right against self-

87

incrimination is not without consequence. The General Assembly may derogate common-law rights by statute, see Holland v. Caviness, 292 Ga. 332, 337 (737 SE2d 669) (2013), which is just what it attempted to do in the DUI context by requiring drivers to submit to chemical tests or face legal consequences for refusing to do so. See, e.g., OCGA §§ 40-5-67.1 (c), (d) (describing the circumstances under which a person who submits, or refuses to submit, to a chemical test will have his or her driver’s license suspended); 40-6- 392 (b) (providing that the results of a chemical analysis of blood- alcohol concentration can give rise to certain inferences in a civil or criminal trial); 40-6-392 (d) (providing that a defendant’s refusal to submit to a chemical analysis can be used as evidence against a criminal defendant). By improperly elevating a common-law right to constitutional status, this Court, in my humble opinion, overstepped its bounds and stripped the General Assembly of its authority to protect the public from dangerous drivers. See Ga. Const. of 1983, Art. III, Sec. VI, Par. I (“The General Assembly shall have the power to make all laws not inconsistent with this

88

Constitution, and not repugnant to the Constitution of the United States, which it shall deem necessary and proper for the welfare of the state.”). Given that the stare decisis factors strongly favor overruling our erroneous self-incrimination precedent, we should correct course, clarify that Paragraph XVI applies only to “testimony,” and relinquish the legislative authority that this Court long ago unconstitutionally assumed.

Accordingly, I would overrule this Court’s precedent in which we have held that Paragraph XVI of Georgia’s 1983 Constitution applies to nontestimonial self-incriminating acts, including Olevik, 302 Ga. 228, Elliott, 305 Ga. 179, and Awad, 313 Ga. 99.34 As a

34 I would also disapprove of this Court’s decisions to the extent that they

stated or implied that Paragraph XVI or prior versions of Georgia’s constitutional right against self-incrimination applied to nontestimonial self- incriminating acts. See, e.g., Dunbar v. State, 309 Ga. 252 (845 SE2d 607) (2020); State v. Turnquest, 305 Ga. 758 (827 SE2d 865) (2019); State v. Herrera- Bustamante, 304 Ga. 259 (818 SE2d 552) (2018); Simpson v. State, 289 Ga. 685 (715 SE2d 142) (2011); Quarterman v. State, 282 Ga. 383 (651 SE2d 32) (2007); Muhammad v. State, 282 Ga. 247 (647 SE2d 560) (2007); Fantasia v. State, 268 Ga. 512 (491 SE2d 318) (1997), overruled on other grounds by Olevik, 302 Ga. 228; Brown v. State, 262 Ga. 833 (426 SE2d 559) (1993); Batton v. State, 260 Ga. 127 (391 SE2d 914) (1990); Green v. State, 260 Ga. 625 (398 SE2d 360) (1990); Thornton, 253 Ga. 524, overruled on other grounds by Neal v. State, 290 Ga. 563 (722 SE2d 765) (2012); Strong v. State, 231 Ga. 514 (202 SE2d 428)

89

result, I would affirm the trial court’s denial of Ammons’s motion to suppress evidence of her refusal to perform a preliminary breath test, the results of her horizontal gaze nystagmus test, and her refusal to perform other field sobriety tests, including the walk-and- turn test.

I am authorized to state that Justice McMillian joins this opinion concurring in part and dissenting in part.

(1973), overruled on other grounds by Williams v. State, 296 Ga. 817 (771 SE2d 373) (2015); Creamer v. State, 229 Ga. 511 (192 SE2d 350) (1972); Manor v. State, 225 Ga. 538 (170 SE2d 290) (1969), vacated in part on other grounds, 408 U. S. 935 (92 SCt 2856, 33 LE2d 750) (1972); Moton v. State, 225 Ga. 401 (169 SE2d 320) (1969); Gunter v. State, 223 Ga. 290 (154 SE2d 608) (1967); Aldrich, 220 Ga. 132; Foster v. State, 213 Ga. 601 (100 SE2d 426) (1957); Thomas v. State, 213 Ga. 237 (98 SE2d 548) (1957); Atterberry v. State, 212 Ga. 778 (95 SE2d 787) (1956); Shepherd v. State, 203 Ga. 635 (47 SE2d 860) (1948); Boyers v. State, 198 Ga. 838 (33 SE2d 251) (1945); McIntyre v. State, 190 Ga. 872 (11 SE2d 5) (1940); Johns v. State, 178 Ga. 676 (173 SE 917) (1934), overruled on other grounds by Corbin v. State, 211 Ga. 400 (86 SE2d 221) (1955); Herndon v. State, 178 Ga. 832 (174 SE 597) (1934); Rawlings v. State, 163 Ga. 406 (136 SE 448) (1926); Groce v. State, 148 Ga. 520 (97 SE 525) (1918); Calhoun, 144 Ga. 679; Elder v. State, 143 Ga. 363 (85 SE 97) (1915); Springer v. State, 121 Ga. 155 (48 SE 907) (1904); Dozier v. State, 107 Ga. 708 (33 SE 418) (1899); Evans, 106 Ga. 519; Myers v. State, 97 Ga. 76 (25 SE 252) (1895); Franklin v. State, 69 Ga. 36 (1882); Blackwell, 67 Ga. 76; Day, 63 Ga. 667.

90

Decided November 2, 2022 — Reconsideration denied November 17, 2022.

DUI. Paulding Superior Court. Before Judge Beavers.

Hunter J. Rodgers, Michael B. Syrop, Dustin S. Townsend, for appellant.

Matthew W. Rollins, District Attorney, A. Brett Williams, Assistant District Attorney; Christopher M. Carr, Attorney General, Patricia B. Attaway Burton, Deputy Attorney General, Paula K. Smith, Senior Assistant Attorney General, for appellee.