delivered the opinion of the Court,

It is well-rooted in our jurisprudence that contempt is a broad and inherent power of a court. But, we have also recognized that despite the breadth and necessity of that power, it is a power that must be exercised with caution. Today, we decide as a matter of first impression whether a trial court may hold a litigant in contempt for perjury committed during a deposition. We are further presented with a question arising from the bifurcated nature of the Texas judiciary and our limited habeas jurisdiction: whether we should exercise our mandamus jurisdiction to provide a forum for a civil litigant who is deprived of liberty pursuant to a court’s contempt order, and the Court of Criminal Appeals has declined to exercise its habeas jurisdiction.

In the underlying civil case, the relator was held in contempt and confined for perjuring himself during a deposition. *363The relator challenged his confinement by seeking a writ of habeas corpus in the Court of Criminal Appeals, but that court declined to exercise its jurisdiction citing, among other things, the civil nature of the case. The Court of Criminal Appeals directed the relator to pursue his remedies in this Court. Because we lack habeas jurisdiction in this case, the relator pursued relief by filing the instant petition for writ of mandamus to challenge his confinement.

We conclude the trial court abused its discretion by holding the relator in contempt for perjury occurring during a deposition, because such perjury did not obstruct the operation of the court. Further, because the underlying suit is civil in nature, and the Court of Criminal Appeals declined to grant the relator leave to file a habeas petition in that court, we hold the relator has no adequate remedy by appeal and therefore mandamus is the appropriate remedy to correct the trial court’s abuse of discretion. We conditionally grant relief.

I. Background

The trial court found relator Coy Reece in contempt for perjuring himself during a 2008 deposition in the underlying civil suit brought by real party in interest SB International, Inc. (SB). Specifically, during the deposition, Reece denied knowledge of a codefendant’s transactions in violation of a contract with SB,1 but later admitted to deliberately making misstatements in the deposition.

After Reece’s admission, SB filed a motion for sanctions, requesting costs and fees incurred in connection with its efforts to obtain Reece’s compliance with discovery obligations, totaling $58,331.17.2 SB secured a hearing on the motion for sanctions, but the trial court suspended the hearing and issued a show cause order to determine if Reece had committed criminal contempt. At the contempt hearing, Reece testified he had lied under oath during the deposition. On May 29, 2009, the court found beyond a reasonable doubt that Reece was guilty of “repeatedly, deliberately, and intentionally [lying] under oath” and that his lying under oath “was reasonably calculated to impede, embarrass, and/or obstruct the court in the discharge of its duties.” The trial court found Reece in contempt and sentenced him to sixty-days confinement, with all but two weeks of the sentence suspended.

Reece filed a petition for writ of habeas corpus with the court of appeals, which granted his request for emergency relief, but that court later dismissed the habeas petition for want of jurisdiction.3 Reece then sought habeas relief from the Court of Criminal Appeals, which denied his initial petition because he was not in custody. On June 24, 2009, after the trial court entered a revised commitment order and Reece was taken back into custody, Reece filed another habeas petition with the *364Court of Criminal Appeals. The Court of Criminal Appeals issued an order denying leave to file the habeas petition without prejudice, explaining that (1) the underlying case is civil in nature, (2) the Texas Supreme Court thus has jurisdiction to entertain Reece’s petition, and (3) although the Court of Criminal Appeals possesses jurisdiction to act on Reece’s petition, it declines to do so in order for Reece to pursue his remedies in the Texas Supreme Court. In re Reece, No. WR-72,199-02, slip op. at 2 (Tex.Crim.App. June 24, 2009) (per curiam) (not designated for publication) (order denying leave to file for habe-as relief).

Reece filed a motion for reconsideration in the Court of Criminal Appeals, explaining that this Court lacks habeas jurisdiction because the contempt order does not emanate from a violation of an order, judgment, or decree. See Tex. Gov’t Code § 22.002(e). Reece also filed the instant mandamus petition with this Court.4 In his mandamus petition, Reece argues that (1) the show cause order is deficient and does not afford him due process, (2) even if the show cause order is not deficient, perjury committed during a deposition is not punishable by contempt, and (3) this Court possesses mandamus jurisdiction to review a contempt judgment involving confinement in a civil case when the Court of Criminal Appeals refuses to exercise its habeas jurisdiction. We granted Reece’s motion for temporary relief and ordered him released on bond; he has served four days of his fourteen-day sentence.

To be entitled to mandamus relief, a relator must demonstrate (1) the trial court clearly abused its discretion, and (2) the relator has no adequate remedy by appeal. In re McAllen Med. Ctr., Inc., 275 S.W.3d 458, 462 (Tex. 2008) (orig. proceeding). We begin by considering the first prong in our mandamus inquiry: whether the trial court clearly abused its discretion by holding Reece in contempt.

II. Clear Abuse of Discretion

The crux of Reece’s arguments rests on his assertion that perjury committed during a deposition is not punishable by contempt, and therefore the trial court clearly abused its discretion by holding him in contempt for this conduct. In evaluating the merits of this assertion, we first examine the general contours of a court’s contempt power in Texas and then decide whether the specific misconduct at issue in this case is punishable by contempt.

A. Contempt of Court in Texas

We have broadly defined contempt as “disobedience to or disrespect of a court by acting in opposition to its authority,” Ex parte Chambers, 898 S.W.2d 257, 259 (Tex. 1995) (orig. proceeding), and observed that contempt is a broad and inherent power of a court, see Ex parte Browne, 543 S.W.2d 82, 86 (Tex. 1976) (orig. proceeding). But, despite the breadth of a court’s contempt power, we have warned it is a tool that should be exercised with caution. Herring v. Houston Nat’l Exch. Bank, 113 Tex. 337, 255 S.W. 1097, 1104 (1923) (orig. proceeding); accord Ex parte Taylor, 807 S.W.2d 746, 748 (Tex.Crim.App. 1991). As the Court of Criminal Appeals has explained, “[cjontempt is strong medicine”— the alleged contemnor’s very liberty is often at stake — and so it should be used “only as a last resort.” Ex parte Pink, 746 S.W.2d 758, 762 (Tex.Crim.App. 1988) (citing Willson v. Johnston, 404 S.W.2d 870, 873 (Tex.Civ.App. — Amarillo 1966, orig. proceeding)).

*365Contempt may occur in the presence of a court (direct contempt), or outside the court’s presence (constructive contempt). Ex parte Gordon, 584 S.W.2d 686, 688 (Tex. 1979) (orig. proceeding). In direct contempt cases, the court must have direct knowledge of the behavior constituting contempt, In re Bell, 894 S.W.2d 119, 127 (Tex.Spec.Ct.Rev. 1995), while the converse is true in a case of constructive contempt. As a result of this distinction, the trial court in a direct contempt proceeding is entitled, in some instances, to conduct a summary proceeding in which the alleged contemnor is not entitled to notice and a hearing, id,5 while a constructive contemnor is always entitled to notice and a hearing in order to defend or explain the charges, see Ex parte Gordon, 584 S.W.2d at 688; Ex parte Werblud, 536 S.W.2d 542, 546 (Tex. 1976) (orig. proceeding) (observing -that constructive contempt entitles the contemnor to more procedural safeguards than those afforded to direct contemnors); see also Ex parte Krupps, 712 S.W.2d 144, 147 (Tex.Crim.App. 1986) (explaining that constructive contempt adjudications satisfy due process if the con-temnor is given notice, a hearing, and the opportunity to obtain an attorney).

Contempt is further classified into either civil or criminal contempt. In determining whether contempt is civil or criminal, it is necessary to examine the purpose behind the contempt order: civil contempt is “remedial and coercive in nature” — the contemnor carries the keys to the jail cell in his or her pocket since the confinement is conditioned on obedience with the court’s order, Ex parte Werblud, 536 S.W.2d at 545,6 while criminal contempt is punitive in nature — “the contem-nor is being punished for some completed act which affronted the dignity and authority of the court,” id. Thus, the distinction between criminal and civil contempt does not turn on whether the underlying litigation is civil or criminal, but rather on the nature of the court’s punishment. Ex parte Chambers, 898 S.W.2d at 266 (Gonzalez, J., dissenting); see generally Ex parte Werblud, 536 S.W.2d at 545-46.

Here, Reece has admitted to lying outside the trial court’s presence during a deposition, and the purpose of the contempt judgment was to punish him for this misdeed rather than to coerce him into complying with an order. The trial court therefore held Reece in constructive criminal contempt.

Reece first asserts that a court’s constructive contempt power is limited to situations in which a person violates a court’s order, and thus Reece cannot be held in contempt here because he did not violate any order. We disagree with this overly narrow definition. While our case law suggests constructive contempt generally arises from the violation of a court order,7 there are situations where a party *366or attorney to a suit could engage in behavior that may warrant a judgment of constructive contempt. For example, in Ex parte Privitt, a criminal defendant was found in contempt for attempting to bribe prospective jurors. 127 Tex.Crim. 475, 77 S.W.2d 663, 664 (1934); see also Ex parte Gordon, 584 S.W.2d at 688 (“Constructive contempt ... is contemptuous conduct outside the presence of the court, such as the failure or refusal to comply with a valid court order”) (emphasis added); Ex parte Murphy, 669 S.W.2d 320, 321 (Tex.Crim.App. 1983) (affirming judgment holding attorney in constructive contempt for failing to attend hearing and trial).8 We decline Reece’s invitation to limit the boundaries of a court’s constructive contempt power in the manner he proposes.

Reece next argues that, even if constructive contempt is not limited to violations of court orders, it is at least confined to acts that obstruct, impede, or embarrass the court in the discharge of its duties. SB responds that although obstruction is a necessary element of contempt under federal law, it is not under Texas law.

At the outset, we acknowledge that federal statutory contempt standards are more stringent than Texas statutory standards. Under federal law, a court has discretion to punish for contempt only to the extent that a contemnor (1) engages in misbehavior in or near the court’s presence so as to obstruct the administration of justice, (2) in the case of a court officer, engages in misbehavior during official transactions, or (3) disobeys or resists a court order. 18 U.S.C. § 401. In Texas, however, the statutory guidance is more sparse. Section 21.002 of the Government Code sets forth the only statutory framework for contempt, broadly providing that a “court may punish for contempt.” Tex. Gov’t Code § 21.002(a).9 Any restrictions on this authority are found in the common law.

Contrary to SB’s assertion, however, it is firmly established in our common law that a court’s ability to punish for contempt is not unfettered. Rather, an act must impede, embarrass, or obstruct the court in the discharge of its duties in order to constitute constructive contempt.10 See Ex parte Norton, 144 Tex. 445, 191 S.W.2d 713, 714 (1946) (orig. proceeding); Ex parte Krupps, 712 S.W.2d at *367149; In re Bell, 894 S.W.2d at 127. “The essence of contempt is the contemn[o]r’s conduct obstructs or tends to obstruct the proper administration of justice.” Lee v. State, 799 S.W.2d 750, 752 (Tex.Crim.App. 1990); see Ex parte Gibson, 811 S.W.2d 594, 596 (Tex.Crim.App. 1991) (per curiam) (setting aside a judgment for contempt arising from an attorney sending an angry letter to the court of appeals because this act did not disrupt the court in the performance of its duties).11 Because conduct must additionally obstruct the court in the performance of its duties to constitute constructive contempt, we turn to whether a litigant’s misstatements during a deposition amount to such an obstruction.

B. Perjury as Grounds for Contempt

This Court has not had occasion to determine whether perjury in a deposition obstructs a court’s performance and thus constitutes contempt. Federal law includes obstruction as an element of contempt and is therefore instructive in our analysis.

It is axiomatic under federal law that perjury alone is not a ground for contempt unless the conduct also obstructs the court in the performance of its duties. See In re Michael, 326 U.S. 224, 228, 66 S.Ct. 78, 90 L.Ed. 30 (1945) (concluding that perjury alone does not constitute an obstruction justifying the exertion of a court’s contempt power); Ex parte Hudgings, 249 U.S. 378, 383, 39 S.Ct. 337, 63 L.Ed. 656 (1919) (same); Matusow v. United States, 229 F.2d 335, 341 (5th Cir. 1956) (“It is well settled that proof of perjury alone will not sustain a conviction for contempt, but misbehavior constituting obstruction of the court must also be established.”). Because a person’s conduct must obstruct a court in the performance of its duties to constitute constructive contempt, we hold that perjury committed during a deposition does not alone constitute constructive contempt. To constitute constructive contempt, such perjury must actually obstruct a court in the performance of its duties.

SB argues that even if obstruction is a necessary element in a constructive contempt case, Reece’s perjury met this element because it interfered with and prejudiced SB during the underlying litigation. While we agree that Reece’s perjury undoubtedly may have caused SB difficulty in the discovery process, we cannot say, on this record, that it obstructed the court in the performance of its duties. Reece’s behavior was reprehensible: he admitted to making intentional misstatements during his deposition concerning core issues in the underlying litigation, behavior which inevitably caused his opponent delays and additional costs. See Soliz v. State, 97 S.W.3d 137, 142 (Tex.Crim.App. 2003) (observing that perjury, including in a deposition, has the effect of hindering the accurate resolution of legal proceedings). But we fail to see how Reece’s misstatements during his deposition rose to the level of actually obstructing the court in the performance of its duties. Neither the contempt orders, nor the record, demonstrate this additional element of obstruction.

*368During discovery it is inevitable that inconsistencies will arise, but the function of our legal system is to ascertain the truth from within these inconsistencies. See In re Michael, 326 U.S. at 227-28, 66 S.Ct. 78. As the Supreme Court reminds us:

All perjured relevant testimony is at war with justice, since it may produce a judgment not resting on truth. Therefore it cannot be denied that it tends to defeat the sole ultimate objective of a trial. It need not necessarily, however, obstruct or halt the judicial process. For the function of trial is to sift the truth from a mass of contradictory evidence, and to do so the fact finding tribunal must hear both truthful and false witnesses.

Id. We are loath to contemplate a system where litigants and their attorneys scour transcripts, searching for any misstatement for the sole purpose of accusing the opponent of contempt and ultimately securing the opponent’s confinement. Authorizing contempt in a situation such as the instant one also begs several questions. Is any misstatement during a deposition punishable by contempt, no matter how immaterial? Or, as the Supreme Court has noted, could a court, in believing a witness testified falsely, punish the person for contempt with the purpose of compelling the person to testify in a manner the court considers truthful? See Ex parte Hudgings, 249 U.S. at 384, 39 S.Ct. 337. The Supreme Court has warned of “a potentiality of oppression and wrong” that could result from authorizing a court to exercise its contempt power in a situation of stand-alone perjury. See id. If a court may hold a litigant in contempt for making misstatements during a deposition, the risk for the “oppression and wrong” the Supreme Court contemplated becomes too great. See id.

A trial court has other weapons in its arsenal to discourage perjury during a deposition without resorting to restraining a person’s liberty in a contempt proceeding. First, a deposition is a discovery tool. As such, our rules make available the possibility of a range of sanctions for discovery abuse. See Tex.R. Civ. P. 215; Bodnow Corp. v. City of Hondo, 721 S.W.2d 839, 840 (Tex. 1986) (per curiam). Here, SB moved for sanctions in response to Reece’s admission that he lied during the deposition; the trial court issued a show cause order concerning contempt on its own initiative. The trial court could have found that Reece abused the discovery process and imposed sanctions ranging from payment of SB’s expenses caused by the failure, including attorney’s fees, to rendering a default judgment. See Tex.R. Civ. P. 215.3, 215.2(b)(1) — (5), (8).

Second, a trial court has the option of referring a perjury allegation to the district attorney for criminal prosecution. See Soliz, 97 S.W.3d at 146 (noting the power of courts to prosecute perjury within the parties’ depositions). As applicable to this case, a person commits the offense of perjury if, with intent to deceive, the person makes a false statement under oath; such an offense is punishable as a Class A misdemeanor. Tex. Penal Code § 37.02. The greater offense of aggravated perjury is available if the perjury is made in connection with an official proceeding and is material, punishable as a third degree felony. Id. § 37.03.12

*369“Courts ‘must be on guard against confusing offenses to their sensibilities with obstruction to the administration of justice.’ ” Ex parte Gibson, 811 S.W.2d at 596 (quoting Brown v. United States, 356 U.S. 148, 153, 78 S.Ct. 622, 2 L.Ed.2d 589 (1958)). Because we hold that perjury committed during a deposition is only punishable by contempt if the act additionally obstructs the court in the performance of its duties, and that no such obstruction occurred here, the trial court’s contempt orders against Reece are void.13 Accordingly, we conclude the trial court abused its discretion by holding Reece in contempt.14

III. No Adequate Remedy by Appeal

We now turn to whether Reece has met the second element required for the exercise of our mandamus jurisdiction: the lack of an adequate remedy by appeal. The parties agree this Court lacks jurisdiction to grant habeas relief in this matter; we may only exercise our habeas jurisdiction when the contemnor’s confinement is on account of a violation of an order, judgment, or decree previously made in a civil case. See Tex. Gov’t Code § 22.002(e); Tex. Const, art. V, § 3. The Court of Criminal Appeals — and not this Court — is the state’s appellate court afforded general original habeas jurisdiction.15 See Tex. Const. art. V, § 5; Ex parte Thompson, 273 S.W.3d 177, 181 (Tex.Crim.App. 2008). Reece contends that because the Court of Criminal Appeals refuses to exercise its habeas jurisdiction, he is left with no avenue to challenge his confinement. Under these circumstances, Reece argues this Court should exercise its mandamus jurisdiction and release him from confinement. SB counters that this Court lacks jurisdiction over Reece’s mandamus petition because it is inherently a habeas petition styled as a mandamus petition: SB contends habeas is the only vehicle for challenging a person’s confinement in a contempt case, and a party may not relabel a petition as one of mandamus to evade this requirement.

As stated above, the Texas Constitution grants this Court limited jurisdiction to review lower court decisions through the writ of habeas corpus. Tex. Const. art. V, § 3(a) (“The Supreme Court *370and the Justices thereof shall have power to issue writs of habeas corpus, as may be prescribed by law.... ”). Section 22.002(e) of the Government Code prescribes the reach of this Court’s habeas jurisdiction:

The supreme court or a justice of the supreme court, either in termtime or vacation, may issue a writ of habeas corpus when a person is restrained in his liberty by virtue of an order, process, or commitment issued by a court or judge on account of the violation of an order, judgment, or decree previously made, rendered, or entered by the court or judge in a civil case.

Tex. Gov’t Code § 22.002(e) (emphasis added). Thus, if the basis of contempt is not a previous violation of a court order, this Court lacks jurisdiction to review a sentence of confinement by way of writ of habeas corpus, even in a civil case. Ex parte Morris, 162 Tex. 530, 349 S.W.2d 99, 100 (1961) (orig. proceeding).

We have previously indicated that a petition for writ of habeas corpus is generally the only method for attacking an order of contempt. See Deramus v. Thornton, 160 Tex. 494, 333 S.W.2d 824, 827 (1960) (orig. proceeding) (“We have uniformly held in this State ... that the validity of a contempt judgment can be attacked only collaterally and that by way of habeas corpus.”). However, we also left open the possibility of circumstances “where the writ of habeas corpus would not be adequate and where mandamus would be the proper remedy.” Id.; see also Dunn v. Street, 938 S.W.2d 33, 35 (Tex. 1997) (orig. proceeding) (per curiam) (citing Deramus and observing same).

On this basis, we have held mandamus is available to challenge an order of contempt not involving confinement given the unavailability of the Court’s statutory habeas corpus jurisdiction in that circumstance. See In re Long, 984 S.W.2d 623, 625 (Tex. 1999) (orig. proceeding) (per curiam) (“Contempt orders that do not involve confinement cannot be reviewed by writ of habeas corpus, and the only possible relief is a writ of mandamus.”); Rosser v. Squier, 902 S.W.2d 962, 962 (Tex. 1995) (orig. proceeding) (per curiam) (entertaining the exercise of mandamus jurisdiction to determine whether fine-only contempt order was void, given the unavailability of the Court’s habeas jurisdiction). But, today, we examine whether section 22.002 permits this Court to exercise its mandamus jurisdiction when the contemnor is confined.

Like SB, several courts of appeals have presumed mandamus is limited to the review of fine-only contempt orders, but not orders that result in confinement.16 In determining whether this maxim is correct, we first consider the contours of both our own and the Court of Criminal Appeals’ habeas jurisdiction, and then address our mandamus jurisdiction.

A. Habeas Corpus Jurisdiction

Before determining whether mandamus relief is available here, we note, and the parties do not dispute, that Reece may not *371pursue relief through the writ of habeas corpus in this Court. By virtue of the particularities of this state’s bifurcated judicial system and the problems arising from the division of habeas corpus jurisdiction between the Court of Criminal Appeals and this Court, we lack habeas jurisdiction over this case. The Court of Criminal Appeals is the court of last resort for criminal matters, Tex. Const, art. V, § 5, while this Court is the court of final review for civil matters, id. art. V, § 3. This framework, while at times imperfect, has been in place since 1876, and is the cornerstone of the bifurcated system of appeals in this state. See State ex rel. McNamara, Co. Atty. v. Clark, 79 Tex.Crim. 559, 187 S.W. 760, 762 (1915) (“Whether wisely or unwisely, the people of this state in framing their Constitution divided the jurisdiction of the civil and criminal courts of final resort, conferring on the Supreme Court final jurisdiction in all civil matters, and upon [the Court of Criminal Appeals] final jurisdiction in all criminal matters.”).

The history of the two high courts’ respective habeas jurisdictions is less than clear. Until 1876, this Court possessed the general power to issue writs of habeas corpus, as well as appellate jurisdiction over criminal law matters. Tex. Const. of 1869, art. V, § 3; Tex. Const. of 1866, art. IV, § 3; Tex. Const, of 1861, art. IV, § 3; Tex. Const, of 1845, art. IV, § 3; see also Ex parte Jackson, 113 Tex. 58, 252 S.W. 149, 150 (1923) (orig. proceeding). But in 1876, the newly adopted Texas Constitution removed jurisdiction over criminal eases from this Court and conferred such jurisdiction to the newly created Court of Appeals, which was granted not only jurisdiction over criminal appeals and certain low-dollar civil appeals, but also general authority to issue writs of habeas corpus. Tex. Const. art. V, § 3 (amended 1891 and 1980); id. § 6 (amended 1891); see Ex parte Jackson, 252 S.W. at 150.17 At the same time, the Constitution revoked this Court’s habeas corpus jurisdiction. Tex. Const. art. V, § 3 (amended 1891 and 1980). In 1891, the Texas Constitution was amended to abolish the Court of Appeals and create the Court of Criminal Appeals, limiting the Court of Criminal Appeals’ appellate jurisdiction to criminal law matters, and granting this Court habe-as corpus jurisdiction once more, but only as “may be prescribed by law.” Tex. S.J. Res. 16, 22nd Leg., R.S., 1891 Tex. Gen. Laws 199, 199-200 (amending Tex. Const. art. V, §§ 3, 5). This constitutional provision has remained unchanged for over 100 years.

Though the Texas Constitution granted this Court habeas jurisdiction as “may be prescribed by law,” we lacked any actual statutory authority to grant writs of habe-as corpus until 1905, when the Legislature enacted Senate Bill 36. This bill authorized us to exercise habeas jurisdiction, but only in situations where a contemnor’s confinement is by virtue of a violation of an order in a civil case. Act of February 24, 1905, 29th Leg., R.S., ch. 17, § 1,1905 Tex. Gen. Laws 20, 20 (current version at Tex. Gov’t Code § 22.002(e)). This statute has remained substantively the same since its enactment and continues in effect today.

As a matter of historical context, the Legislature apparently enacted Senate Bill 36 to ensure the Court of Criminal Appeals determines criminal matters and this Court civil matters in habeas proceedings, in line with the bifurcated system contemplated in our Constitution. As the full *372report from the House Judiciary Committee states:

Under the present status of the law all writs of habeas corpus sued out by a person who has been fined or imprisoned for a violation of any order, judgment or decree entered in a civil case, go upon appeal to the Court of Criminal Appeals instead of the Supreme Court. This often results in confusion and conflicts of decision authority. The object of the bill is to harmonize our judicial decisions and jurisdiction so that all civil matters shall be determined by the Supreme Court and all criminal matters by the Court of Criminal Appeals.

House Judiciary Comm., Full Report, Tex. S.B. 36, 29th Leg., R.S. (1905) (emphasis added). Over the past 100 years, contempt orders arising from civil cases concern a contemnor’s violation of a court order the vast majority of the time. See James L. Branton, Note, Ex parte Morris, 162 Tex. 530, 349 S.W.2d 99 (1961), 40 Tex. L.Rev. 413, 416 (1962) (noting the rarity of contempt judgments arising from civil cases that do not involve violations of a prior court order and recommending a legislative amendment to grant this Court jurisdiction over all contempt matters arising from civil cases). The Legislature, perhaps, simply did not envision contempt in civil cases extending beyond this limited scope, and so crafted this Court’s habeas jurisdiction accordingly. Similarly, the Court of Criminal Appeals has suggested the same purpose behind Senate Bill 36: to confer original habeas corpus jurisdiction to this Court in all civil cases in order to better maintain the bifurcated court system contemplated in the Texas Constitution. See Ex parte Wolf, 116 Tex.Crim. 127, 34 S.W.2d 277, 278 (1930) (noting that the Court of Criminal Appeals should avoid exercising habeas jurisdiction in civil cases due to the possibility of “find[ing] ourselves in conflict with the civil courts, and thus interfer[ing] with [the civil courts’] proper functioning and the administration of justice”); Ex parte Zuccaro, 72 Tex.Crim. 214, 162 S.W. 844, 845 (1913) (observing that the Court of Criminal Appeals has jurisdiction in criminal cases and has no jurisdiction in “civil cases of any character”).

Nonetheless, this Court has generally properly declined to exercise habeas jurisdiction in contempt cases arising from civil suits where a violation of a court order is not the grounds for contempt, as required by the plain language of section 22.002(e) and its predecessors.18 But the result of *373this Court’s limited habeas jurisdiction in civil cases has led to an occasional inefficient back-and-forth between the Court of Criminal Appeals and this Court on the issue of jurisdiction, as well as some difficulties when the underlying case is civil in nature, but falls inside the statutory loophole created by the particular division of habeas jurisdiction between the Court of Criminal Appeals and this Court.

Despite its general original habeas jurisdiction, the Court of Criminal Appeals has deferred to this Court in cases where the two courts clearly have concurrent habeas jurisdiction over civil matters.19 And, as in the instant case, the Court of Criminal Appeals has preferred to defer to this Court in contempt proceedings arising from civil cases before exercising its habe-as jurisdiction, even when it is less evident this Court possesses jurisdiction over the matter.20 The issue we must resolve is whether mandamus is an appropriate remedy in a situation where we lack habeas jurisdiction to review a contempt order involving confinement arising from a civil case, but the Court of Criminal Appeals declines to exercise its general habeas jurisdiction.

B. Mandamus Jurisdiction

Unlike our habeas jurisdiction, our constitutional and statutory grant of mandamus jurisdiction is broad; this Court possesses general original jurisdiction to issue writs of mandamus. See Tex. Const, art. V, § 3(a) (granting the Court *374power to issue writs of mandamus as specified by the Legislature); Tex. Gov’t Code § 22.002(a) (permitting the Court to issue writs of mandamus “agreeable to the principles of law regulating those writs”). Because we ascertain nothing in the statutory grant of our mandamus jurisdiction precluding us from granting relief here, and further determine the circumstances of this case warrant the exercise of our mandamus jurisdiction, we conclude mandamus is an appropriate remedy for Reece’s unlawful confinement.

First, the limit SB proposes on our mandamus jurisdiction is unwarranted as a matter of statutory construction. While a person must be confined for us to exercise our habeas jurisdiction, this is so only because the habeas statute specifically states that a person must be “restrained in his liberty” for that remedy to be available. See Tex. Gov’t Code § 22.002(e). But the converse is not true: our statutory grant of habeas jurisdiction does not prohibit us from exercising mandamus jurisdiction when a person is confined. See id. Nor does the statute prescribing our mandamus jurisdiction state that mandamus relief is unavailable when a person is confined — instead, as noted above, our statutory grant of mandamus jurisdiction is broad. Id. § 22.002(a). The Legislature could have included language designating habeas corpus as the exclusive remedy for unlawful confinement, but declined to do so. Cf. MCI Sales & Serv., Inc. v. Hinton, 329 S.W.3d 475, 504 (Tex. 2010) (declining to limit a statutory designation as exclusive since the terms of the statute did not include such a limitation). We will not read such a limitation into the statutory scheme governing our mandamus jurisdiction.

Further, mandamus is an “extraordinary remedy, not issued as a matter of right, but at the'discretion of the court.” In re Prudential Ins. Co. of Am., 148 S.W.3d 124, 138 (Tex. 2004) (orig. proceeding). “Mandamus review of significant rulings in exceptional cases may be essential to preserve important substantive and procedural rights from impairment or loss....” Id. at 136. Mandamus is a remedy not restricted by “rigid rules” that are “necessarily inconsistent with the flexibility that is the remedy’s principle virtue.” Id.; see also In re McAllen, 275 S.W.3d at 464 (noting that whether a clear abuse of discretion can be remedied on appeal “depends heavily on circumstances, it must be guided by analysis of principles rather than simple rules that treat cases as categories”). And mandamus is a proper vehicle for this Court to correct blatant injustice that otherwise would elude review by the appellate courts. See In re Prudential, 148 S.W.3d at 138.

It is difficult to imagine a circumstance more compelling for the exercise of our mandamus jurisdiction than a situation where the relator’s liberty interests are threatened without a remaining procedural safeguard for challenging his confinement. Certainly, a result of our inaction is a potential waste of judicial and litigant resources as the case travels between this Court and the Court of Criminal Appeals, with neither court exercising jurisdiction to consider the merits of Reece’s petition. But the further consequence of Reece’s lack of an adequate appellate remedy in this matter is his unlawful confinement: the relator’s very liberty interests are at stake with no other procedure to challenge his confinement in our state courts.

Finally, because mandamus is expressly reserved for situations where a relator lacks an adequate remedy by appeal, we have employed it in other situations where, as here, a statutory gap prevents the relator from obtaining adequate appellate relief. For example, before the Legislature *375enacted section 51.016 of the Civil Practice and Remedies Code, a party improperly denied the benefit of arbitration under the Federal Arbitration Act had no right to interlocutory appeal in state courts. See Jack B. Anglin Co. v. Tipps, 842 S.W.2d 266, 272 (Tex. 1992) (orig. proceeding). The Court therefore held mandamus was the appropriate remedy in wrongful denials of motions to compel arbitration under the FAA since otherwise the very subject of the appeal — the right to arbitrate, as contracted for by the parties — would be rendered illusory. Id. Eventually, of course, the Legislature closed this gap by enacting a law authorizing interlocutory appeals under the FAA in Texas courts. See Tex. Civ. Prao. & Rem.Code § 51.016.21

More relevant to the instant dispute, until 1978, the Court of Criminal Appeals lacked jurisdiction to grant writs of mandamus except as necessary to protect its own judgments. See Tex. Const, art. V, § 5 (amended 1977 and 1980); Thomas v. Stevenson, 561 S.W.2d 845, 847 (Tex.Crim.App. 1978). To fill this jurisdictional gap, this Court occasionally entertained mandamus petitions filed by criminal defendants, primarily involving requests to compel courts to begin trial in accordance with the defendant’s right to a speedy trial. See, e.g., Fariss v. Tipps, 463 S.W.2d 176, 180-81 (Tex. 1971) (orig. proceeding) (observing that this Court has mandamus jurisdiction in civil and criminal cases and defendant seeking speedy trial has no adequate remedy by appeal); Lawrence v. State, 412 S.W.2d 40, 40 (Tex. 1967) (orig. proceeding) (per curiam); Cooper v. State, 400 S.W.2d 890, 890-92 (Tex. 1966) (orig. proceeding); Wilson v. Bowman, 381 S.W.2d 320, 321 (Tex. 1964) (orig. proceeding). In 1978, the people amended the Texas Constitution, granting the Court of Criminal Appeals mandamus jurisdiction over all criminal law matters. See Tex. S.J. Res. 18, 65th Leg., R.S., 1977 Tex. Gen. Laws 3359, 3359-60; Stevenson, 561 S.W.2d at 84647.22 This Court has since limited its exercise of mandamus review to civil matters. See Stevenson, 561 S.W.2d at 847.23

Our Constitution grants the Court of Criminal Appeals jurisdiction over criminal matters and this Court jurisdiction over civil cases. Tex. Const, art. V, §§ 3, 5. As it has done historically, the Court of Criminal Appeals prudently deferred to this Court because the underlying case is civil in nature. But, by virtue of a statutory limitation, we lack jurisdiction to issue a writ of habeas corpus to discharge the relator from unlawful confinement in this *376civil case. The constitutional and statutory contours of our mandamus jurisdiction, however, are not so limited. Because the Court of Criminal Appeals declines to exercise jurisdiction over this civil case, Reece is left without a procedure for challenging his confinement in our state appellate courts. Accordingly, we hold Reece has no adequate remedy by appeal. We note that our holding is limited to situations where the underlying dispute is civil in nature, and the Court of Criminal Appeals declines to exercise its habeas jurisdiction. A relator does not lack an adequate remedy by appeal when the Court of Criminal Appeals entertains the merits of a habeas petition but dismisses with prejudice or denies relief. In other words, a relator is not entitled to a “second bite at the apple” by virtue of our holding.

C. Response to Dissent

In his dissent, Justice Willett argues that an “express statutory prohibition” precludes our exercise of mandamus jurisdiction here. Justice Willett points to the unavailability of our habeas jurisdiction in this matter as a “sign” that we are not meant to hear appeals of any sort — including by way of mandamus — arising from contempt orders not involving a violation of a previous court order. As such, Justice Willett reads the statute permitting us to issue writs of mandamus “agreeable to the principles of law regulating those writs,” Tex. Gov’t Code § 22.002(a), as circumscribed by the limitations in our habeas statute. Yet, the plain text of the statute governing our original proceeding jurisdiction does not comport with Justice Wil-lett’s interpretation. As discussed above, neither our statutory grant of habeas or mandamus jurisdiction prohibits us from exercising mandamus jurisdiction when a person is confined, and we decline to read such a limitation into the language of the statute.

Despite his statements to the contrary, the rule Justice Willett’s dissent attempts to establish is that express legislative permission is required for us to exercise our mandamus jurisdiction. Under his reasoning, if a statute grants jurisdiction in only a limited circumstance, it must follow that we are forbidden from exercising our mandamus jurisdiction in a situation falling outside the parameters of that limitation. The weakness inherent in this argument is obvious in the context of arbitration. We could have interpreted the Legislature’s specific allowance for interlocutory appeal under the Texas Arbitration Act in the absence of a law allowing for the same under the Federal Arbitration Act as a “sign” that the Legislature did not intend for any appeal, including by mandamus, under the FAA. But we wisely declined to limit our mandamus jurisdiction so narrowly.

Further, a review of our precedent reveals its support for our — and not Justice Willett’s — interpretation of our mandamus jurisdiction. We decline to overrule Deramus because the general proposition in Deramus is still valid. In Deramus, we stated that a petition for writ of habeas corpus was the only manner for challenging a contempt judgment, but left open the possibility of the availability of mandamus in a situation where habeas would prove inadequate. See Deramus, 333 S.W.2d at 827. Seizing on that exception, we have exercised our mandamus jurisdiction in situations of fine-only contempt where, under the constraints of the habeas statute, we lacked habeas jurisdiction. See Rosser, 902 S.W.2d at 962; In re Long, 984 S.W.2d at 625; Tex. Gov’t Code § 22.002(e) (authorizing this Court to issue a writ of habeas corpus “when a person is restrained in his liberty ” by virtue of violating a previous order) (emphasis added). Here, as in Ros-ser and In re Long, the habeas statute *377does not permit our exercise of habeas jurisdiction, but it does not follow that we may not exercise our broader mandamus jurisdiction. In Rosser and In re Long, we declined to read the limitations in our ha-beas statute as a legislative prohibition against our exercise of mandamus jurisdiction. We decline to do so here as well.

We finally observe that whatever one’s view of our bifurcated judicial system, it is the system established by the people of this state. The bifurcated system, pursuant to our Constitution, establishes this Court as the court of final resort for civil matters, see Tex. Const, art. V, § 3, and the Court of Criminal Appeals as the court of final resort for criminal matters, see id. art. V, § 5. The instant case — a business dispute in which a party to the case was held in contempt for lying during a deposition — is unquestionably civil, and not criminal, in nature. A case of criminal contempt does not depend on whether the underlying case is civil or criminal, see generally Ex parte Werblud, 536 S.W.2d at 545-46, nor require the same procedural safeguards as those required in a criminal case, see id. at 547 (noting that contempt statute is exempted from requirements of a jury trial). Under our bifurcated system, it is legal, as well as logical, for this Court to exercise its mandamus jurisdiction in a civil case where the Court of Criminal Appeals has refused to exercise its jurisdiction and has deferred to this Court. While a man’s very liberty is at stake, Justice Willett focuses on policy issues concerning our bifurcated judicial system that are beyond the purview of the case before us today and are best left to the people of Texas and their elected legislative officials. The people have entrusted this Court with expertise over civil matters and our sister court with the same over criminal matters. We honor this decision in exercising our jurisdiction over this civil case today.

IV. Conclusion

Because the trial court abused its discretion in confining Reece for criminal contempt for acts of perjury occurring during a deposition and Reece has no adequate remedy by appeal, we conditionally grant the writ of mandamus and order the trial court to vacate the May 29, 2009, and June 24, 2009, contempt judgments against Reece. We are confident the trial court will comply, and our writ will issue only if it does not.

dissenting.

The Court of Criminal Appeals has general original habeas jurisdiction, Tex. Const. art. V, § 5; Ex parte Thompson, 273 S.W.3d 177, 181 (Tex.Crim.App. 2008), while this Court’s habeas jurisdiction is limited. Our habeas jurisdiction exists in matters where a contemnor is confined because he or she violated “an order, judgment, or decree previously made, rendered, or entered by the court or judge in a civil case.” Tex. Gov’t Code § 22.002(e).

Although our habeas jurisdiction is limited, our mandamus jurisdiction is broad. See Tex. Const, art. V, § 3(a); Tex. Gov’t Code § 22.002(a). And for the reasons the Court sets out, I agree that our broad mandamus jurisdiction encompasses the matters set out in Reece’s petition. Nevertheless, and as SB International, Inc. argues, Reece substantively petitions this Court for habeas relief. Under the circumstances I would refrain from granting mandamus relief for the reasons Justice *378Willett sets out in part IV of his dissent, which I join.

Because I disagree that we should grant mandamus relief, I respectfully dissent.

joined by Justice JOHNSON as to Part IV, dissenting.

Intrepidity at the Alamo; entering the United States as the Republic of Texas; fifty-eight Texas-born recipients of the Medal of Honor; Bob Wills and George Strait; Nolan Ryan and Babe Didrikson Zaharias; five Super Bowl titles (sadly none this millennium); Dr Pepper and the “little creamery” in Brenham; deep-fried anything at the State Fair; a spirit of daring and rugged independence — the sources of Lone Star pride are innumerable.

Unfortunately, the juris-imprudent design of the Texas judiciary does not make the list. Today’s case is a byproduct of that recondite web, sparking a game of jurisdictional hot potato between us and our constitutional twin, the Court of Criminal Appeals. Truth be told — and this particular truth has been told repeatedly — the State’s entire Rube Goldberg-designed ju-. dicial “system” is beyond piecemeal repair; it should be scrapped and rebuilt top-to-bottom. That said, and however labyrinthine the jurisdictional maze often is, the answer in today’s case seems straightforward: This dispute belongs with our sister court. It arrived on our doorstep because of a simple yet pivotal misunderstanding: the Court of Criminal Appeals’ mistaken belief that we have unfettered habeas jurisdiction and are thus equally able to grant habeas relief.1 We do not,2 and the Court today is unified 9-0 on that point (though the Court does not explicitly mention our sister court’s misinterpretation). We part ways 7-2 on whether we should make lemonade out of jurisdictional lemons by wiring around our habeas limitation and relabeling the relief sought “mandamus.”

The mandamus remedy turns on two findings: legality and practicality.3 On both scores, I would return this case to the court that conceded two years ago that it “does have the authority to act in this case.”4 Statute and precedent strongly *379suggest we cannot hear this case, but even if we can, practical considerations advise we should not. Neither refusing nor resisting, the Court today yanks tighter a Gordian knot that should be cut clean through. I respectfully dissent, and, for good measure, exhort the Legislature to propose a judiciary worthy of Texas.

I. This Case Illustrates (Again) Our Abstruse Judicial “System.”

“An artificial and arbitrary system, as age creeps on, gets hardened arteries.”5

The history of Texas courts is indeed a sclerotic one. But all’s well that ends well, and even a helter-skelter judicial structure might be worth the strife if it still managed, despite itself, to produce efficiency. Ours hasn’t.

A. We Have Arrived Here Through Historical Happenstance.

Like many things of Texas lore, the story of our court system begins with its size. During the colonization of Texas, judicial power was vested in the “municipal alcalde, an elected official who held executive, legislative and judicial duties,”6 and Stephen F. Austin was himself the court of last resort.7 After winning our independence, “Texas began with a unified judiciary system,”8 and both the Republic of Texas and the early State had a single high court with both civil and criminal jurisdiction.9 From statehood through Reconstruction, every appeal from a trial court went directly to the state’s Supreme Court, which at varying times had three or five members.10

“Forty years and five constitutions later,” 11 and responding to this Court’s congested docket, the Constitution of 1876 created a three-judge court of appeals for criminal matters and limited this Court’s jurisdiction to civil matters.12 The court of appeals, which despite its name was not an intermediate court, had final say in criminal appeals, and could also hear civil matters involving less than $1,000.13 The flow *380of cases continued unabated, however, and in 1879, the Legislature fashioned another judicial Band-Aid with the creation of a Commission of Appeals.14 But even doubling the number of commissioners provided scant docket relief, and in 1891 (just fifteen years after the Constitution was adopted), the citizens of Texas tried another approach, a massive overhaul that scrapped the entire Judiciary Article of the Constitution.15 This kitchen-sink reform abolished the court of appeals and Commission of Appeals, gave criminal jurisdiction to a new Court of Criminal Appeals, and created three new intermediate courts of civil appeals16 (in Galveston, Forth Worth, and Austin).17 Our Court would maintain its civil-only docket and focus chiefly on resolving conflicts in the courts of appeals.18

The Legislature was also charged with the task of dividing the state into judicial districts, each with its own' court of civil appeals.19 In 1913, this Court’s jurisdiction grew to include all cases from the courts of civil appeals,20 and in 1980, a constitutional amendment bestowed criminal jurisdiction on the renamed courts of appeal.21 Efforts to create a separate body of criminal-only intermediate courts were defeated.22 So while the two highest courts in the state maintain specialized dockets, the feeder courts beneath them do not.

Generally speaking, under our bifurcated structure, litigants file civil matters in the Supreme Court and criminal matters in the Court of Criminal Appeals. People frequently get misdirected, though — lawyers included — and the courts’ front offices regularly redirect lost litigants to the “other” high court. In fact, this Court’s clerk’s office has a stock letter it sends — every single day — to lost litigants, steering them to our sister court and noting that the Supreme Court “does not have jurisdiction over criminal cases” and “does not review the decisions of the Court of Criminal Appeals.” 23

Our dual high courts are largely meant to be co-equals — constitutional twins. This is anomalous among court systems, even in the only other two-court state, Oklahoma. Like Texas, Oklahoma has a Supreme Court that hears civil appeals and a Court of Criminal Appeals that *381hears criminal appeals.24 But there are two key differences. First, the Oklahoma Court of Criminal Appeals is “subject to the power of the Legislature to change or abolish.”25 Second, the Oklahoma Supreme Court is truly supreme; if there is a jurisdictional clash, the Supreme Court “shall determine which court has jurisdiction and such determination shall be final.”26 In other words, there are two states in the nation with two courts of last resort. But only one state — the Lone Star State — has a non-supreme Supreme Court.27

B. Our Fragmented Structure is Much Maligned, and Deservedly So.

The convoluted make-up of the Texas judiciary — “one of the most complex in the United States, if not the world”28 — does not lack for critics, from the litigants who endure it, the lawyers who navigate it, and the judges who lead it. In 1991, this Court’s appointed Citizens’ Commission on the Texas Judicial System reached a stark but unsurprising conclusion: “Texas has no uniform judicial framework to guarantee the just, prompt and efficient disposition of a litigant’s complaint.... With the passage of time, the organization of the courts has become more, not less cumbersome.”29 That critique mirrors one that same year from the Texas Research League (“TRL”), which former Chief Justice Phillips had asked to scrutinize our judicial structure and suggest concrete improvements. The system’s mind-numbing complexity led TRL to lament in May 1991 that the Texas judiciary was in “disarray” and “ill-equipped to meet the needs of the 21st century,” adding, “Texas does not have a court system in the real sense of the word.”30 Indeed, “assigning the appellation ‘system’ to our state courts might require a long stretch of the imagination.” 31 Nothing has improved, and interestingly, the most strenuous critics, it *382seems, are those who know the system best: the judges.

First, trial courts. “Texas has some 3,241 trial courts within its 268,580 square miles.”32 The complexity at the lower-court level is dizzying, as the attached chart (meant to simplify things) illustrates.33 In his 2007 State of the Judiciary address, Chief Justioe Jefferson urged the Legislature to modernize our patchwork trial-court system, calling on lawmakers to start “examining whether Texans are best served by the current (and often redundant) complex system of county courts at law, district courts and statutory probate courts, or whether streamlining some of these courts may create a simpler system.”34 Three members of this Court recently branded our jurisdictional mishmash “unimaginably abstruse,” a tangle that has “gone from elaborate to Byzantine.” 35 A former member of this Court politely called our system “the opposite of a coordinated judiciary.”36 One former state appellate judge bemoaned our “maze of jurisdiction and procedure” that “[ojnly a puzzle-maker could appreciate.”37

In 1993, the Court-appointed Citizens’ Commission on the Texas Judicial System commented that “[n]o one person understands or can hope to understand all the nuances and intricacies of Texas’ thousands of trial courts.”38 Yet another report bemoaned that “current judicial districts are so fundamentally unfair and so irrationally configured as to shock the conscience of all Texans who familiarize themselves with the present system.”39 This long-derided irrationality persists.

As one might imagine, our bizarre structure has generated some fanciful factoids — practical problems and offbeat jurisdictional oddities that clog the everyday inner workings of our judiciary. Consider:

• Texas has at least nine different types of trial courts, “although that number does not even hint at the complexities of the constitutional provisions and statutes that delineate jurisdiction of those courts.”40 Whether a given trial court has jurisdiction is a five-step inquiry.41

*383• As Chief Justioe Jefferson has pointed out: “Some counties share a multi-county district court, while others have multiple districts within the county. And some counties are part of more than one district, creating a shifting target for litigants who may not know which court’s rules prevail. Overlapping geographical jurisdiction creates confusion for litigants and increases the risk of conflicting rulings in a single area.”42

• At least one county court has no civil jurisdiction whatsoever.43

• Only eight percent of Texas’s justices of the peace are lawyers, even though they can hear cases involving multimillion-dollar claims.44

• A civil suit that would be tried before a twelve-person jury in district court would be tried before a six-person jury if filed in a county court.45

• District court vacancies are filled by appointment by the Governor46 but statutory county court vacancies are filled by appointment by the county commissioners, even though those courts frequently have jurisdiction over the same matters.47

• Whether there is a minimum monetary limit on the State’s district court jurisdiction actually remains an open question.48 While the Constitution has been amended to eliminate a monetary minimum, there is some argument that it is still implied49

• Generally, jurisdictional limits on statutory county courts range widely by county — from $500 to $100,00050 — and some such courts have no monetary limits at all.51

• “Appellate rights can vary depending on which court a case is filed in, even among trial courts with concurrent jurisdiction, and even when the same judge in the same courtroom presides over two distinct courts.”52

Second, intermediate appellate courts. Texas is the only state in the nation in which trial judges answer to more than one intermediate appellate court;53 that is, *384no other state has overlapping appellate jurisdictions.54 Fifteen counties are in overlapping districts.55 This Court has lamented the “manifest” problems inherent in overlapping districts: “uncertainty from conflicting legal authority,” “the potential for unfair forum shopping,” and “jurisdictional conflicts.”56 In fact, the two Houston-based courts of appeals have even reached polar-opposite outcomes on the same facts57 — allowing three passengers in a car accident to sue but not the fourth.58 The following year, in 2002, we exhorted the Legislature that “[n]o county should be in more than one appellate district.”59 I suspect we will do so again next year when we issue our required plan to the Legislature on whether any appellate courts should be added, eliminated, consolidated, or reallocated.60

The Attorney General’s current chief legal counsel recently bemoaned problems inherent in our overlapping intermediate-court structure: “Much of the problem— and most of the opportunity for reform— lies in the antiquated structure of the lower courts”;61 likewise, our appellate courts “are struggling to overcome a structure ill-suited to modern caseloads.”62 Created to ease high-court docket congestion, our “heavily fractured intermediate court structure,” if anything, has created a system “more primed for generating conflicts” than any other state in the nation.63

Third, courts of last resort. Coy Reece’s case is but one more cautionary Texas tale. As it illustrates, our dichotom-ized system invites inter-court confusion, and as Texas history shows, inter-court clashes. The Citizens’ Commission report from 1993 noted that conflicts between the dual courts have arisen over the conclusivity of the courts of appeals’ factual determinations, the constitutionality of the “Pool Hall Law,” and whether journals of the House and Senate can be used to contradict an enrolled bill.64

In fact, members of the two courts have themselves sometimes highlighted the friction that occasionally befalls a bifurcated system. A Court of Criminal Appeals judge once lamented the split-system’s tendency to shuffle parties needlessly about as he sent an “appellant on his way to begin yet another search for the proper *385forum.”65 In another case, three members stated that they were “concerned that this State’s bifurcated judicial process could sometimes generate conflicting decisions at the highest level on identical questions of law.... If there is a problem, it lies with the lines dividing the constitutional jurisdiction of this Court and the Texas Supreme Court.”66 The Texas system’s decentralized nature has been blamed for a “lack of coordination”67 that is apparent here. Even the Office of the Attorney General — “the law firm of Texas” itself — is not wholly immune from the jurisdictional confusion. In 1992, the Attorney General’s Office took the rare step of appealing a lower-court ruling striking down the State’s anti-sodomy law to both courts because, as the lead attorney explained, “We want to make sure we’re not locked out of an appeal. It was either file with both or roll the dice.”68 The Court of Criminal Appeals declined jurisdiction,69 and this Court eventually ruled that it too had no jurisdiction.70 The (non)decision was roundly criticized. One might wonder, as did an editorial board, “What’s the point of having not one, but two final state appellate courts if neither of them has the authority to rule on the constitutionality of a Texas criminal statute?”71 Lawyers ought not be forced to litigate “on a guess and a gamble.”72

Up north in Oklahoma, that Supreme Court could decide this jurisdictional quandary swiftly. Not so here, though one court-reform study, mindful of the potential for jurisdictional confusion, proposed a Sooner-like solution whereby “the supreme court should determine which court has jurisdiction, and those determinations should be final.”73

C. A Century of Pleas for Structural Reform Have Failed.

The urgency of sweeping judicial reorganization was “a perennial theme”74 throughout the twentieth century. Earnest reformers like Roscoe Pound75 and *386blue-ribbon studies galore urged a sweeping restructuring of our hodgepodge judiciary. Throughout the 1900s, “in virtually every decade of [the] century,”76 there were regular calls in the Legislature, the academy, and the profession for structural reforms at every level, including high-court merger.77 There have been periodic small-bore reforms, yet even those piecemeal tweaks were “inexorably tedious and protracted”;78 ad hoc is the rule — evolutionary rather than revolutionary.

The 1970s were particularly reform-minded. The Judicial Section of the State Bar of Texas pushed for substantial changes to our judicial structure during the 1971 legislative session.79 That same year, the Legislature proposed a constitutional amendment, eventually adopted by voters in 1972, directing the Legislature to form a Constitutional Revision Commission to “study the need for constitutional change” and then convene in 1974 as a constitutional convention.80 Also that same year, in October 1971, then-Chief Justice Calvert formed the Chief Justice’s Task Force for Court Improvement to rewrite Article V, the Judiciary Article of the Texas Constitution. In September 1972 the Task Force proposed, among other things, simplifying the trial-court maze, investing the courts of civil appeals with criminal jurisdiction (which happily happened in 1980), reforming judicial selection, and merging our twin high courts.81 The Calvert Task Force coincided with a court-reorganization report by the House Judiciary Committee, which in 1972 called for extensive changes in the judicial *387branch.82

In early 1973, the thirty-seven members of the Texas Constitutional Revision Commission began nine months of study and public hearings, culminating in a proposed new state constitution.83 (The Revision Commission was chaired by then-former Chief Justice Calvert, who had left the Court the previous October, one month after his Task Force unveiled its proposed Judiciary Article). Essentially, the Calvert-led Revision Commission adopted the recommendations of the Calvert-led Task Force.84 Notably, though, the Revision Commission, unlike the Task Force, wrestled with modernizing the entire Texas Constitution, not just Article V. And the document it presented to the Legislature in November 1973 was the first comprehensive effort to draft a new constitution for Texas since the Constitutional Convention of 1875.85

The following January, the Legislature convened unicamerally in the House chamber as the Constitutional Convention of 1974. Like the Revision Commission, the Constitutional Convention favored a wholesale overhaul of the entire Constitution, and many of the proposed reforms, especially a right-to-work provision, provoked raucous debate.86 The Convention dissolved seven months later, falling three votes shy of submitting a new constitution to Texas voters.87 That October, the House Judiciary Committee submitted a report calling on the Legislature to submit to voters the revision of Article V that the 1974 Constitutional Convention considered.88

The Legislature reconvened in January 1975, and this time, acting as a regular legislature and not as a constitutional convention, it approved what became a package of eight separate amendments, including a new Article V, which resurrected the recommendations for a combined high court, courts of appeals with both civil and criminal jurisdiction, and substantial trial-court unification.89 For the first time in a century, Texans had an opportunity to consider a revised constitution. It was not to be. As in the Constitutional Convention the previous year, fierce opposition arose over various non-judiciary proposals (like annual legislative sessions, a right-to-work provision, and taxation and education reforms) and each and every proposed revision was defeated, including the modernized Article V (which received more votes than any other amendment).90

*388A 1976 interim study of the House Judiciary Committee submitted fifteen piecemeal recommendations,91 six of which the Legislature enacted (like the creation of the Office of Court Administration).92 In 1979, then-Chief Justice Greenhill championed in his State of the Judiciary address the rifle-shot reform of giving criminal jurisdiction to the courts of civil appeals,93 and voters agreed in 1980.94

The call for broader reforms persisted throughout the 1990s — from TRL,95 to the Comptroller,96 to the Court-appointed Citizens’ Commission.97 In May 1991, TRL urged a totally new Judicial Article, saying our courts are so “fragmented” that “[t]he Texas court system really is not a system at all.”98 In 1991, we directed an eighty-four-member Citizens’ Commission on the Texas Judicial System to “study and recommend any necessary or desirable improvements in the courts of Texas.”99 Given our constitutional responsibility “for the efficient administration of the judicial branch,”100 the Court invited commonsense reforms, predominantly those related to the “jurisdiction and title of the trial and appellate courts of Texas.”101 Believing “a sound organizational and administrative structure is essential to a well-regarded judiciary,” the Commission proposed a system that simplified general-jurisdiction trial courts and unified our dual high courts, though the new Supreme Court would have “two divisions, civil and criminal, each with seven justices.”102

In the 1990s, the Citizens’ Commission proposals did draw support as part of broader efforts to streamline our ungainly constitution down to something approaching comprehensibility.103 No such luck; the efforts sputtered. Our unwieldy constitution lives, including our crazy-quilt court system, a top-to-bottom mess. The push for modernization has continued apace in the 2000s. Many observers, including members of this Court,104 have *389continued pushing for lower-court simplification, and other voices urge high-court merger as part of a broader restructuring.105

Against this bizarre background I turn to Reece’s petition for writ of habeas corpus. It determines the procedural posture that so interestingly animates this case, and channels the kinds of cases this Court can and cannot hear. The issue of jurisdiction — deciding to decide — may sound like a meta-interest floating in the jurisprudential ether, but its importance as a threshold issue cannot be overstated. The matter of to whom the courts are open— and for which claims — colors our bifurcated high-court system, and ultimately disposes of this case. Sections II and III discuss, respectively, the statutory and precedential evidence that suggests we are not permitted by law to hear this case. Section IV explains that even if we do maintain jurisdiction, it would be unwise to exercise it. The former is a matter of a legal directive, the latter a matter of judicial discretion, but both yield the same conclusion: There is no compelling case to hear this case.

II. The Clear Statutory Prohibition that Prevents this Court from Hearing this Case as a Habeas Petition Suggests it Cannot be Cleverly Restyled as Mandamus.

There is no argument that this Court is statutorily hamstrung when it comes to habeas jurisdiction. The Texas Constitution gives us the “power to issue writs of habeas corpus, as may be prescribed by law.” 106 That law is Section 22.002(e) of the Government Code, which limits such jurisdiction to times “when a person is restrained in his liberty by virtue of an order, process, or commitment issued by a court or judge on account of the violation of an order, judgment, or decree previously made, rendered, or entered by the court or judge in a civil case.”107 Despite the jurisdictional thicket that has sprouted kudzu-like around us, the path out is rather linear.

A. There is No Debate that Habeas May Not Issue Here.

As applies here, this Court has the authority to issue a habeas writ only if Reece both seeks release from custody and appeals from an order of contempt based on a violation of an order, judgment, or decree “previously made” by the court or judge in a civil case.108 Otherwise, we *390have no statutory authority to act: If the basis for contempt is not the violation of a previously issued order, we do not have jurisdiction to review a sentence of confinement via habeas corpus.109 While Reece does seek release from custody, there is no argument — either by the Court,110 the trial court below, or the very parties before us — that the contempt order here was based on such a violation. Therefore, this Court lacks power to issue habeas relief.

To say this is a rule grounded in statute and in precedent would be an understatement. This Court has been in the business of reviewing habeas petitions based on statutory language similar to Section 22.002(e) for more than 100 years.111 We have denied jurisdiction over habeas petitions not arising from the violation of a previously made order for just as long.112 The Court of Criminal Appeals was, respectfully, incorrect when it stated that “[ejffective 1981, Article 5, § 3(a) of the Texas Constitution was amended to give the Texas Supreme Court and the Justices thereof the authority to issue writs of ha-beas corpus.”113 The amendments of 1980 (effective 1981) did no such thing. Instead, they rewrote the first paragraph of the section, but retained the language that “[t]he Supreme Court and the Justices thereof shall have power to issue writs of habeas corpus, as may be prescribed by law ....”114 And here, law prescribes that habeas may not issue in this case.

*391B. There Should Be No Debate that Relabeling the Remedy “Mandamus” Cannot Circumvent this Rule.

The strength of this rule should be heeded as a sign — both from the Texas Legislature and our many decisions construing its enactments — that we are not meant to hear appeals from contempt cases where the basis for contempt is not the violation of a previously issued order, and therefore, we are not meant to ad-lib the means to arrive at the same forbidden end. It is undisputed that we cannot hear this case as a habeas petition. Why, then, should we be permitted to hear it under another name?

This statutory prohibition — the only legislatively mandated anchor in a sea of confused and overlapping jurisdiction — should and does provide a comprehensive sense of this case. By issuing mandamus when we are clearly not permitted to issue habeas, we do a disservice to the framework differentiating the two, as well as to the jurisdictional structure which (for better or worse) we are charged with upholding.

The Court contends “our constitutional and statutory grant of mandamus jurisdiction is broad,” and not limited in the way I suggest here, explaining that “this Court possesses general original jurisdiction to issue writs of mandamus.”115 But the Court is forced to qualify that proposition by citing to the Texas Constitution: “See Tex. Const, art. V, § 3(a) (granting the Court power to issue writs of mandamus as specified by the Legislature).”116 The exception nullifies the rule. Our mandamus jurisdiction is undoubtedly circumscribed by law. Where the Legislature has spoken clearly and removed the kind of case now before us from our jurisdiction, it is disingenuous to circumvent the rule by renaming the remedy.

III. Precedent Further Indicates this Court Cannot Issue Mandamus.

The statutory prohibition against habeas is but one reason to dismiss the case. There are others grounded in our mandamus (rather than habeas) jurisprudence. We have at least suggested — if not stated plainly — that the habeas prohibition precludes our ability to hear a ease like this, explaining that “[o]ur original habeas corpus jurisdiction is limited thereby to cases in which a person has been confined for violating an order, judgment or decree in a civil cause, and we are without power to inquire into the legality of restraint imposed for some other reason.”117

A. Our Deramus Decision Demonstrates — Rather Than Disproves— that Habeas is Inappropriate Here.

Both Reece and the Court118 rely heavily upon one sentence in Deramus v. Thornton, in which we preserved the possibility that there might be contempt-related situations where mandamus, not habe-as, would be the proper remedy: “We are *392not to be understood as saying, however, that there may not arise conditions involved in contempt matters where the writ of habeas corpus would not be adequate and where mandamus would be the proper remedy.”119 The cautious words of wise jurists intent on protecting a hypothetical situation, however, should not be read to apply to and permit any series of facts that follow.

Deramus was held in contempt for violating an injunction. He sought a writ of mandamus ordering his trial judge to vacate the contempt judgment and dismiss the contempt proceedings. There, as here, we noted that the usual avenue for such a situation was habeas, explaining: “Had the District Judge not suspended the judgment of contempt the normal course would have followed, and the remedy adopted by the relator could necessarily have been an application for a writ of habeas corpus.”120 Contrary to Reece’s urgings, in Deramus we noted that mandamus was not proper, buttressed in large part by the reasoning that “[w]e have uniformly held in this State ... that the validity of a contempt judgment can be attacked only collaterally and that by way of habeas corpus.”121 Even when we noted that this question was arguably a matter of policy — a view that seems to pervade the Court’s looser approach to the issue — we still remained “reluctant to depart from a judicial path so well landmarked, especially so since the claimed inadequacy of habeas corpus ... [was] one common to all cases where escape is sought from the penalties of a contempt judgment.”122 We went on to explain that “[t]his in itself, we think, is sufficient justification for our refusal of this application. To do otherwise would completely change the procedure long followed in this State and allow in every case an attack on the order of contempt by way of mandamus.” 123

Reece has not violated a previously issued order, meaning this Court may not issue habeas. If the Court in Deramus found that habeas — not mandamus — was appropriate where contempt was the result of the violation of a previously made order, then the Court should find here that neither habeas — nor mandamus — is proper where there is no such violation. The dissent in Deramus acknowledged that ha-beas was the usual remedy in cases where a relator seeks a release from confinement, but maintained the view that the import of the cases suggested a different principle: “[W]here a judge, as in the instant case, has determined to commit and fine a relator on a void contempt judgment, this court has the power to issue writs of mandamus and prohibition to prevent the enforcement of a void act.”124 It is unclear why the Court has essentially taken up the Deramus dissent without explicitly overruling Deramus.

In sum, what was true in Deramus remains true today: Granting mandamus “would completely change the procedure long followed in this State and allow in every case an attack on the order of con*393tempt by way of mandamus.”125 Deramus imagined scenarios in which the inadequacy of habeas would render mandamus the proper route. But Deramus itself demonstrates that this is not such a scenario.

B. In re Long Indicates Mandamus Specifically May Not Issue Here, Where Contempt Sanctions Involve Confinement.

In re Long126 reinforces this conclusion. There we held mandamus would be proper in the review of contempt sanctions not involving confinement.127 But because Reece challenges a criminal contempt action involving confinement, not based upon a previously issued order, In re Long is not directly controlling. Therefore, Reece fits into neither category. The Court’s assertion that because we have “declined to read the limitations in our habeas statute as a legislative prohibition against our exercise of mandamus jurisdiction” in fine-only cases, we can rightly “decline to do so here as well”128 is a non sequitur. Reece was confined, so cases about non-confined persons cannot support the leap made by the Court today. There is simply no precedent establishing that mandamus is the appropriate remedy in this case.

In fact, quite the opposite is true. It can be inferred from In re Long that contempt sanctions that do involve confinement may not be reviewed through mandamus — otherwise, the distinction that case makes would be meaningless. We have applied this kind of logic to the habeas cases discussed earlier, in which we reasoned from the statute permitting issuance of the habeas writ where there is a violation of a previous court order that we were prohibited from issuing the writ where there was no such order. It makes sense to do the same here. Under a simple corollary of the In re Long rule, we are prohibited from issuing mandamus because Reece was subject to contempt sanctions that involved confinement. Even foregoing this inference as the Court would, however, it is clear that mandamus has never been permitted where there was no violation of an order and confinement was involved.

C. The Court Misconstrues the Mandamus Remedy.

The Court’s defense of its decision is based largely on four contentions: (1) mandamus is generally flexible; (2) no law announces that habeas is the exclusive remedy; (3) mandamus has often been used to “gap-fill” where there is no remedy; and (4) our sister court tends to defer to us on matters such as these. The first two stem from a more general view about the mandamus and habeas remedies, respectively. The second two are rooted in case law. I address each in turn.

First, it is true, as the Court points out, that mandamus is available to review rulings in “exceptional cases,” and that “rigid rules ... are necessarily inconsistent with the flexibility that is the remedy’s principal virtue.”129 But we have regularly deferred to the Legislature’s determinations *394of when mandamus is appropriate.130 And the remedy has been largely used in obviously civil cases with no criminal element, and generally has not been used to trump other, independent limitations that work to bar the mandamus remedy.131 This case is distinguishable on both counts: It presents an underlying civil case with a criminal penalty and is independently limited by precedent that confines the mandamus remedy when it comes to criminal contempt.

Consider, as an illustration, Betts v. Johnson,132 There, the Court determined that an independent statutory limitation prevented it from issuing mandamus, and consequently overruled a motion to file a mandamus petition.133 The statutory limitation was an article that permitted the Court to issue mandamus against an “officer of the state government.” 134 But since the writ applied for was against a board of officers, not against an officer, the Court reasoned that it could not issue mandamus.135 The text of the statute prevented it from doing so. As recently as 2001, some justices of this Court refused mandamus on the same logic — not as a matter of practicality, but instead as a matter of legality.136

Similarly, we have determined mandamus may not issue to controvert a prior injunction137 or to compel an officer to act outside the bounds of the law.138 In both *395instances, the existence of law that would conflict with the mandamus remedy functions as an independent limitation upon it. We have also found ourselves powerless to issue mandamus where a collection of statutes suggested we lacked jurisdiction to hear the case in the first place.139 While this is, to be sure, different from a prohibition specifically against mandamus, it still reinforces the general rule that independent statutory limitations channel the mandamus power.

Second, it is also true that no statutory or constitutional provision states that ha-beas is the only vehicle in this circumstance, and this Court has previously granted mandamus relief in quasi-criminal cases. The first point is a cat’s game, in which neither side wins: While there may not be a provision limiting the possibilities in this case to habeas, there is certainly not one explicitly permitting mandamus. To the contrary, In re Long at least suggests that mandamus is inappropriate where confinement is involved.140 This and the fact that there is a statutory provision in Section 22.002(e) specifically limiting this Court’s habeas jurisdiction to certain instances — none of which are presented here — are at least two thumbs on the scale for the view that this Court lacks jurisdiction in this instance. As to the second point, having granted mandamus relief in other quasi-criminal cases does not make it appropriate here. Neither party seems to be able to point to a case where this Court granted mandamus on facts such as these, and in the teeth of an independent statute and precedent circumscribing our ability to do so.

Third, according to the Court, many cases support the view that mandamus is a statutory “gap-filler” that may issue here. But in none of these cases did an independent statutory prohibition suggest that mandamus was inappropriate as it does here. Instead, we simply fashioned a remedy because there was a lack of available alternatives. In other words, we are not claiming that express statutory permission is required to issue mandamus; we are simply asking that the Court refrain from issuing mandamus where there seems to be an express statutory prohibition against doing so.

Contrary to the Court’s understanding of my position, I do not believe that “if a statute grants jurisdiction in only a limited circumstance, it must follow that we are forbidden from exercising our mandamus jurisdiction in a situation falling outside the parameters of that limitation.” 141 We have certainly gap-filled properly in the past. This case is distinguishable because there is no gap (the Court of Criminal Appeals can still act, as Reece’s motion for rehearing remains pending there) and because there is statutory evidence not only that our jurisdiction is limited, but also that our jurisdiction is explicitly prohibited in this context.

“Gap-filling” is just that — a decision to act when there is no other avenue open to the parties. Today the Court does something more akin to “needle-threading” than “gap-filling.” It attempts to gap-fill where there is not clearly a gap — where there is, instead, a law.

*396For this reason, the Court’s citations to various arbitration cases are off-point. In Jack B. Anglin Co. v. Tipps,142 the Court determined that mandamus was appropriate to review a trial court’s denial of a motion to compel arbitration under the Federal Arbitration Act (“FAA”).143 But because of an independent statutory limitation — Texas procedural rules — those claiming a right to arbitration under the Texas Arbitration Act (“TAA”) and alternatively the FAA were required to file both an interlocutory appeal under the TAA and a writ of mandamus under the FAA.144 Was this dual requirement a model of efficiency? Clearly not. But the Court recognized then — as it should realize now — that the form of the remedy was circumscribed by legislative mandate.145 We reaffirmed that principle in the next case, too.146

We gap-filled because without the mandamus remedy the essence of the appeal would vanish, and the arbitration-seeker would be left without the very thing for which he contracted ex ante.147 But we did not ignore law suggesting or stating we could not do so. Here, there is a wealth of law tending to show the mandamus remedy is not permitted. There is not a gaping hole that suggests oversight, but a narrow cranny that suggests deliberation. In other words, despite the Court’s argument, a “specific allowance for interlocutory appeal under the TAA in the absence of a law allowing for the same under the FAA”148 is not the same as the presence of an independent statute explicitly suggesting interlocutory appeal is not allowed under the FAA. I doubt we would have permitted mandamus if there had been such a limitation.

This Court has similarly stepped in149 where the Court of Criminal Appeals could not issue mandamus except as necessary to protect its own judgments.150 This line of cases only reinforces my own view. We have granted mandamus where the Court of Criminal Appeals was constitutionally prohibited from doing so. We should not short-circuit the Court of Criminal Appeals from issuing habeas where we are statutorily prohibited from doing so.

It is also worth noting that the Legislature eventually intervened to fill in the *397“gap” for each of these cases.151 That pattern only demonstrates that if the Legislature determines that the failure to give our Court jurisdiction over cases such as these was mere oversight, it knows well how to correct the error. Unless and until it does so, however, it makes little sense to take the lid off this jurisdictional can of worms, particularly when the can belongs to another.

Fourth and finally, the Court claims that the Court of Criminal Appeals has “preferred to defer” to this Court where contempt proceedings arise from civil cases. In support, it points to a case in which the Court of Criminal Appeals, after doing so, was forced to take the case back after a Supreme Court justice explained that this Court did not have jurisdiction.152 That example should be followed today. The essence of this question is not in how often the Court of Criminal Appeals may err in attempting to pass a case like this to us, but in how often we have erred in accepting it. That we have never done.

Were we the Oklahoma Supreme Court, sorting out jurisdictional spats like this one by simply taking the case might be a more tenable position. But in Texas, it is the Legislature that designs and divvies up the dockets. And the Legislature has not given us the authority to hear this case— whether we call it habeas or mandamus.

D. The Court’s Reliance on Legislative History is Both Unnecessary and Unwise.

My skepticism of legislative history is well known, and well informed. It is a wariness borne of many years participating in the legislative process at both the state and federal levels, and confirmed by six years on the bench, where I see firsthand the perils of “embarking on a scavenger hunt for extratextual clues prone to contrivance.”153 Any imagined gains from *398rummaging around in legislative minutiae, particularly absent any textual ambiguity, are more than dwarfed by multiple realities.

One such reality, unfortunate but also undeniable, is that legislative history is prone to manipulation (by lawyers, judges, and legislators alike) and often cited inaccurately, selectively and misleadingly. More fundamentally, the statute alone is what constitutes the Legislature’s collective will, and isolated snippets along the way lack the authoritative imprimatur of a Legislature that, we must presume, intended precisely what it enacted.

That said, one need not necessarily subscribe to this view to find the Court’s reliance on legislative history unsettling. For this case demonstrates yet another disadvantage to reading through the often-distorting lens of legislative history: It is really no aid at all.

The Court attempts to guess at what the Legislature of 1905 could have meant. The Court’s determination to wrestle with the ghost of Section 22.002(e) reveals, perhaps not surprisingly, that wrestling with ghosts is unsatisfying. After reading the House Judiciary Committee’s report on Senate Bill 36, the Court can only suggest that “perhaps [the Legislature] simply did not envision contempt in civil cases extending beyond [a contemnor’s violation of a court order], and so crafted this Court’s habeas jurisdiction accordingly.”154 It can only note that the Court of Criminal Appeals has “suggested the same purpose.”155

That shaky assumption is the basis for the Court’s assertion that Reece’s case has “fall[en] inside the statutory loophole created by the particular division of habeas jurisdiction between the Court of Criminal Appeals and this Court.”156 This is a loophole, of course, but only if one looks beyond the text of the statute — and not clearly even then. As another matter, “loopholes” are usually passageways through which unaddressed matters threaten to escape. Here, the text of Section 22.002(e) limits habeas jurisdiction to exceedingly specific instances with the kind of precision that suggests the Legislature was drawing lines, not holes. If it wasn’t, then the Legislature remains free to clarify matters — especially in a case that springs from a judicial maze that lawmakers are best positioned to simplify.

IV. Granting Relief Poses Few Practical Benefits and Many Potential Practical Burdens.

Even if we can issue mandamus here, it is doubly clear that we should not. Even if it is legal, it is certainly impractical. Most peculiar about the Court’s decision to accept this appeal is the lack of any compelling practical reason to do so. The Court of Criminal Appeals itself acknowledged two years ago — before the case even arrived on our doorstep — that it has jurisdiction to hear the case.157 If we dismiss, *399Reece will return to our sister court, where a motion for reconsideration remains pending-presumably awaiting our action. This renders untrue the Court’s statement that Reece has “no other procedure to challenge his confinement in our state courts”158 and “no adequate remedy by appeal.”159 The Court of Criminal Appeals has not refused to act; it has instead deferred final action until we act first.160 I agree that “mandamus is a proper vehicle for this Court to correct blatant injustice that otherwise would elude review by the appellate courts,”161 but that scenario simply does not exist here. As discussed above, the Court has utilized mandamus as a flexible remedy only where all other meaningful roads were blocked. If adequate appellate relief is available elsewhere, mandamus should not be used as judicial duct tape to cover “gaps” that simply do not exist. Where our sister court has conceded its own authority to act, we should be doubly disinclined to intervene.

A. Hearing Reece’s Case Implies a Lack of Equal Sisterhood with Our Supposed Sister Court.

If the instant case offers no reason to seize jurisdiction, the specter of future cases more strongly militates against our doing so. Hearing the case implies that the Supreme Court and the Court of Criminal Appeals are not co-equals. Accepting mandamus jurisdiction when the Court of Criminal Appeals has exclusive habeas jurisdiction over these types of contempt orders would violate the mandated separation with that court. It would evince more respect for an institution we call our equal, and those who created it, to allow it to hear its rightful docket than to encroach pointlessly upon it.

B. Hearing Reece’s Case Will Disorient Deciders, Confusing the Two High Courts and Courts of Appeals Alike.

Similarly, this case leaves open the question of whether and when a petitioner may seek review in both courts, and in what order. Such confusion could lead to an unnecessarily increased docket in either court, or at least wasted resources spent shuffling cases between the two systems (or discussing whether to do the shuffle in the first place). While the Court seems concerned that dismissing Reece’s case would constitute “a potential waste of judicial and litigant resources as the case travels between [both courts], with neither court exercising jurisdiction to consider the merits of Reece’s petition,”162 that small, one-time shuffle will save us far more than it will cost. Ignoring the reality that our jurisdiction is limited will only make the ping pong match longer, and with more balls in the air.

The confusion caused in hearing this case will affect both litigants and the courts of appeals below that have understood and applied for years the rule that today the Court contravenes. The court *400of appeals in the instant case certainly believed it was following that rule when it dismissed Reece’s habeas petition for want of jurisdiction. It would be strange to tell those courts that a relator need only style his petition as mandamus to merit jurisdiction, especially when — as the Court acknowledges 163 — those courts have been operating under the assumption that mandamus was the only proper remedy.164 Even Reece himself assumed this was the rule, as indicated by his decision to file “a motion for reconsideration in the Court of Criminal Appeals, explaining that this Court lacks habeas jurisdiction because the contempt order does not emanate from a violation of an order, judgment, or decree.” 165

Further, this issue is before us largely because of the Court of Criminal Appeals’ mistaken view that this Court has habeas jurisdiction.166 Making it a policy to grant cases that arise out of error instead of correcting the error will make neither our court nor our sister court as careful or as diligent in reviewing cases as we ought to be; it will only encourage punting cases— likely, the most difficult cases deserving of the most attention — back and forth between us. Reece claims that these are “unique circumstances” warranting mandamus as a matter of policy; but, contradictorily, Reece also warns that “the next case might involve a party to a civil case sentenced to six months in jail for contempt.” It certainly might. But that would be a matter for the Court of Criminal Appeals.

C. Hearing Reece’s Case May Manufacture Manipulation.

A lack of jurisdictional clarity threatens to encourage forum shopping. An astute attorney may determine that his client stands to receive a more favorable ruling at one court rather than the other, and arrange jurisdiction-manipulative arguments accordingly. After this case, for example, petitioners seeking an audience with the Supreme Court would be advised to style them petitions as mandamus; with the Court of Criminal Appeals, habeas.

D. Hearing Reece’s Case Draws a Blurry, Rather than Bright-Line, Rule.

Finally, the Court’s suggestion that today’s decision draws helpful lines of clarity between the civil and the criminal is unpersuasive. Today the Court may ostensibly limit itself by allowing this Court to grant mandamus in “situations where the underlying dispute is civil in nature, and the Court of Criminal Appeals declines to *401exercise its habeas jurisdiction.”167 This has the semblance of a bright-line rule— we hear appeals arising from underlying civil matters, and the Court of Criminal Appeals from underlying criminal matters. But that is misleadingly simplistic, and contravenes prior precedent in which we explained that case categorization does not dispose of this issue.

In other words, it is still true that “[u]n-der the provisions of the Texas Constitution and the pertinent Texas statutes relating to the original jurisdiction of this Court and the Court of Criminal Appeals, the circumstance that the cause out of which a restraint of a person’s liberty arises may be classified as a civil case, is not sufficient to vest this Court with habeas corpus jurisdiction.” 168 Further, hearing this case, and perhaps future cases like it, may force us to handle appeals from civil cases with criminal penalties, and force us at least in part to take on quasi-criminal matters. An unnecessary, duplicative upsurge in this Court’s docket is alarming enough on its own; but one comprised of quasi-criminal cases when there is a separate court designated for criminal matters is even more insupportable.

It is easy to draw a line based upon the nature of the underlying case, but this case alone demonstrates that the line is somewhat meaningless. Even the Court acknowledges that “the distinction between criminal and civil contempt does not turn on whether the underlying litigation is civil or criminal, but rather on the nature of the court’s punishment.”169 The instant case, then, demonstrates that the cases — like the courts that hear them — are not always cleanly bifurcated. Here, the Court finds constructive criminal contempt even though it was imposed in a civil trial and not for the violation of a court order.170

The world is not nearly as tidy as the approach the Court has designed for it. In those blurry instances, it seems best to follow precedent. And a bright-line rule is particularly without its usual benefits where it comes at the expense of precedent that prohibits such a rule. “[W]e are seldom presented with the opportunity to give a jurisdictional statute a reasonable construction that results in more uniformity and simplicity (even if only slightly more), and given that opportunity in this case, I would seize it.”171 The simplest— and most defensible — approach is not to attempt to create a bright-line rule but to refrain from hearing the case whatsoever. The brightest line is the one drawn between these two courts. After all, the Court of Criminal Appeals has called its original jurisdiction to issue writs of habe-as corpus “unlimited.”172

V. Conclusion

No amount of head-tilting and eye-squinting can manufacture jurisdiction where there is none. Where mandamus relief would mirror the effect of a statutorily prohibited habeas writ, we should not hear the case. We should be particularly hesitant where our own mandamus jurisprudence precludes such relief. We should be triply wary where our sister court concedes it “does have the authority to act in this case.”173

*402I understand the Court’s commendable desire to correct an erroneous trial-court ruling, but where our labyrinthine judicial structure curbs our ability to hear certain cases, we must obey that limitation. This is not a case where either — or neither— high court has jurisdiction. This case belongs at the Court of Criminal Appeals, and that court is apparently awaiting our decision before ruling on Reece’s motion for rehearing.

It makes little sense for us to expand, without clear delineations, our own jurisdiction where our sister high court has already declared it has the power to take action. By inventing jurisdiction without practical or legal reasons for doing so, the Court today further muddles the two-court system in which we find ourselves. We may have inherited a jurisdictional house of cards, but it is imprudent to build more intricate towers upon it.

I would dismiss this petition, and in doing so urge my own: The fastest growing state in the nation requires a modernized top-to-bottom judicial structure fit for the twenty-first century and worthy of our great State. At the very least (and it grieves me to use these six words) Texas should be more like Oklahoma, where one high court is truly supreme and empowered to decide jurisdictional squabbles inherent in a bifurcated scheme. I respectfully dissent.

APPENDIX

*403[[Image here]]