T h e o • p h i l o g u e

Home » 2008 » September (Page 2)

Monthly Archives: September 2008

Brothers, Let us Embrace Heresy!

I am through critiquing Luther’s doctrine of baptism.  Now I am going to draw some implications from our post series that will conclude something that in the prima facie will certainly seem quite absurd … namely, that we should embrace heresy.  How would I ever come to such a conclusion?  Well … the short of it is this.  The Reformed Traditions (at least in our day) have set up Luther as having recovered the gospel in his understanding of sola fide, and anything that contradicts sola fide is considered heresy to many protestants.  The problem is, sola fide must be defined historically according to what the major Reformers actually taught in their doctrine of sola fide.  As we have seen in our series on Luther’s view of baptismal regeneration / justification by baptism / sacramental mediation, Luther’s sola fide should be rejected.   

In short, if orthodoxy if defined by the Reformers, then I’d rather be a heretic (given Luther’s soteriology).     

Luther’s arguments reveal something potentially shocking about his understanding of grace, salvation and faith which have important implications for how we understand the reformation slogan sola fide (justification by faith alone) which has been accredited to Luther.  Although it is far from the scope of this paper to present Luther’s doctrine of justification, a fair summary of it can be given.  Luther supposedly believed that justification was a forensic declaration  in which a sinner is declared to be righteous on the formal basis of an alien righteousness through the instrumentality of faith.   

The part of sola fide which needs to be re-examined in Luther’s theology is the concept of instrumentality.  Many Reformed traditions held faith to be the sole instrumental cause of justification.  That is, one is justified by faith alone—only faith and nothing else.  Luther is mistakenly thought to be the champion of this sola fide doctrine which is thought to be wholly disjunctive with any “Roman Catholic” view of sacramental mediation of saving grace.  

For example, in a relatively recent treatment (2001) of doctrine throughout church history, John D. Hannah misrepresents Luther as believing in sola fide in such a way as to rule out sacramental mediation of saving grace.[1]  His misunderstanding is rooted in a misinterpretation of Luther’s phraseology of baptism as God’s Word.  Since Luther denies that water all by itself saves, but rather asserts salvation through the Word which is attached to the water and faith which receives it, Hannah concludes that Luther did not believe in the efficacy of the sacrament of baptism.  “The sacraments, then, have a subjective function as a witness to faith in God’s generosity; they do not have an objective function of being the actual means of acquiring God’s grace.”[2]  However, as we have seen, a quick overview of Luther’s teaching in the catechism reveals that when Luther emphasizes God’s Word in baptism, he does not have in mind the gospel per say, and faith which receives the gospel before baptism.  Rather, he has in mind water baptism as God’s promise of salvation (God’s Word of promise) and faith in that promise.  Although this is clear from what we have already observed, the following quote from The Babylonian Captivity makes this connection in Luther’s mind more obvious.

Thus you see how rich a Christian is, that is, one who has been baptized! Even if he would, he could not lose his salvation, however much he sinned, unless he refused to believe.  For no sin can condemn him save unbelief alone.  All other sins, so long as the faith in God’s promise made in baptism returns or remains, are immediately blotted out through that same faith, or rather through the truth of God, because he cannot deny himself if you confess him and faithfully cling to him in his promise.  But as for contrition, confession of sins, and satisfaction, along with all those carefully devised exercises of man: if you rely on them and neglect this truth of God, they will suddenly fail you and leave you more wretched than before.  For whatever is done without faith in God’s truth is vanity of vanities and vexation of spirit.[3] [emphasis mine]    

The “truth of God” in this passage is God’s promise to save through baptism.  “Unbelief” is unbelief in this promise.  Perseverance is maintained only for those who “faithfully cling to him in his promise” [emphasis mine].  Hannah seems to be completely unaware of Luther’s basic paradigm for baptism as water comprehended in God’s Word.  It is hard to understand, if Hannah has read Luther on baptism, how he could possibly miss Luther’s constant emphasis on baptism as the cause of all saving grace, and therefore the cause of justification.[4]  Although Luther did not believe that baptism could save unless faith is present, with faith present (whether before or after baptism), the sacraments are “‘effective in the sense that they certainly and effectively impart grace where faith is unmistakably present.”[5]     

We know that wherever there is a divine promise, there faith is required, and that these two are so necessary to each other that neither can be effective apart from the other.  For it is not possible to believe unless there is a promise, and the promise is not established unless it is believed.  But where these two meet, they give a real and most certain efficacy to the sacraments. … Thus Christ says: “He who believes and is baptized will be saved; but he who does not believe will be condemned”[6] [Mark 16:16, emphasis mine]. 

When Luther says “it is not baptism that justifies or benefits anyone, but it is faith in that word of promise to which baptism is added,”[7] he is often misinterpreted (as with Hannah) as teaching a pure sola fide which rules out baptism as efficacious for salvation.  This is a typical case of Reformed bias reading into Luther’s theology what is not actually there.  The Reformed tradition will have to look elsewhere for a pure sola fide champion hero in the Reformation.  Luther’s view of justification in the context of his theology of baptism can only be viewed as heretical in many Reformed traditions today.  Likewise, the modern Reformed traditions which hold to a sola fide absent of sacramental mediation were considered heretical to Luther,[8] and he considered faith in baptism as salvific as part of saving faith.[9] 

Hannah is also guilty of not reckoning with Luther’s distinction between baptism as a work of God and not a work of human effort, which leads him to conclude that Luther could not have seen baptism as being an instrumental cause of the forgiveness of sins and of the removal of moral inability: “For Luther, water baptism does not cleanse the guilt and inability inherited through original sin.. … Thus, any notion of causative cooperation, even a gracious cooperation, is impossible because humankind has no merit to commend itself to God.”[10]  Luther, however, as we have seen, did not see baptism as man’s work, but God’s work.  Therefore, he did not see baptism as human merit commending itself to God, but as an act which “brings—victory over death and the devil, forgiveness of sin, God’s grace, the entire Christ, and the Holy Spirit with his gifts.”[11]  In these two areas, Hannah reveals his bias by misrepresenting Luther.  

As should be blindingly unambiguous from reading Luther’s arguments for baptismal regeneration and infant baptism, Luther believed in a sacramental mediation of all saving grace, and therefore the grace of justification would also be mediated by the sacrament of baptism.  Luther not only allowed for “works” (baptism) to be an instrumental cause in saving grace, he demanded it with a passion, indicting anyone who opposed him as opposing God’s Word.  Since the slogans of the Reformation are usually attributed to Luther, this has implications for how we understand the doctrine of sola fide in its historical sense.  Is there really a singular “Reformed” position on the doctrine of sola fide?  Do we understand it as the sola fide of Luther, Calvin, Zwingli, or some other Reformer or Reformed Tradition?  This study also has implications for the ect debate over sola fide.  Are those evangelicals who signed the ect documents compromising sola fide because Rome believes in a sacramental mediation of saving grace?  Are those who remain “faithful” to the Reformation the ones who rule out sacramental mediation or demand for sacramental mediation?  It seems that if we are going to define the Reformation by Luther’s doctrine of sola fide, the latter would be the case.[12]

Although the conclusions I have drawn may seem provocative and controversial, they simply flow from a study of Luther’s doctrine of baptism.  Reformed traditions have invested so much energy making Luther their Reformation hero, their ranks who defend sola fide can hardly stand to read Luther for what he actually taught about baptism, and the implications it has for his doctrine of justification.  Instead, they paint him with a biased brush for the sake of ecclesiological and theological expediency.  Luther’s views of baptism and his argumentation for baptismal regeneration and infant baptism demonstrate a lack of hermeneutical discernment in Luther as well as a lack of logical discernment.  While we can credit Luther with many good things, such as his emphasis on faith, repentance, and piety in an age of ritualistic notions of the sacraments that did not include sincere inner spirituality,[13] we should be very careful about defining orthodoxy and heresy based on the so-called “Reformed position.”  If Luther’s position on justification is the orthodox view known as sola fide, then Reformation orthodoxy must be rejected and heresy must be embraced.[14]       



[1] John D. Hannah, Our Legacy: The History of Christian Doctrine (Colorado Springs, CO: NavPress, 2001), 227-229.   

[2] Ibid., 229.  In the same vein, Hannah represents Luther has having a view in which “the symbol has no efficacy.” Ibid.  Lohse tries to correct this false interpretation of Luther’s “sign” language (pardon the pun).  “When Luther at times used the word ‘sign,’ particularly in his doctrine of the Supper, that use may not be construed in Zwinglian terms.  Luther never intended the term to be merely ‘symbolic.'” Lohse, Martin Luther’s Theology, 300.           

[3] Martin Luther, Luther’s Works, vol 36, Word and Sacrament II, 60. 

[4] “Luther thus places baptism in the center of the Christian life.  His understanding of baptism exactly expresses his doctrine of justification.  Through the sacrament of baptism we are ‘sacramentally’ or ‘because of the sacrament,’ made completely pure and innocent in God’s gracious judgment, that is, we are ‘children of grace and justified persons’. … His doctrine of baptism is basically nothing else than his doctrine of justification in concrete form.”  Althaus, The Theology of Martin Luther, 356.  

[5] Martin Luther, Luther’s Works, vol 36, Word and Sacrament II, 67.  

[6] Ibid., 67. 

[7] Ibid., 66. 

[8] Lohse recognizes that Luther accused the Anabaptists, for example, of “works-righteousness and even idolatry.”  Lohse, Martin Luther’s Theology, 305. 

[9] Faith in the Word of God in the context of Luther’s theology of baptism means faith in baptism. “This is why Christ immediately adds, ‘He who does not believe will be condemned’ even though he is baptized, for it is not baptism, but faith in baptism, that saves.”  Althaus, The Theology of Martin Luther, 364.  

[10] Hannah, Our Legacy, 227.   

[11] Luther, The Book of Concord, 442. 

[12] Leonardo De Chirico, however, understands the ECT project to be guilty of a lack of systemic awareness for upholding agreement over justification by faith on the one hand, and baptismal regeneration on the other.  Leonardo De Chirico, “Christian Unity vis-à-vis Roman Catholicism: A Critique of the Evangelicals and Catholics Together dialogue,” Evangelical Review of Theology 27 no. 4 (2003): 346. 

[13] Luther’s doctrine of baptism must be seen as reactionary to prevalent medieval notions of ex opere operato which “tended to ritualize and desiccate baptism.  In essence this Latin formula meant that the sacraments infused grace simply form the use of them, apart from any act of the soul. … Second, in the development of the medieval sacramental system, baptism tended to be associated only with the beginning of life, its chief role being to wash away the guilt of original sin.”  Tranvik, “Luther on Baptism,” 25. 

[14] Another important feature to Luther’s arguments for infant baptism, which I did not have time to focus on, but which is probably almost as important is this: Luther held an extremely high view of tradition.  Althaus points out the fact that although church tradition “is certainly not Luther’s last word on the subject…it is certainly his first.”  Althaus, The Theology of Martin Luther, 359.  In Althaus’ estimation, Luther had a “high evaluation of the universal tradition of the church,” and used similar argumentation on other occasions.  Ibid, 363.

Expect More Anti-Calvinism Rhetoric

From  my limited perspective, it seems to be slowly becoming a fad for churches to do anti-Calvinism Conferences.  For example, the upcoming SBC president Johnny Hunt’s church First Baptist Woodstock recently hosted an anti-Calvinism conference with some pretty big wig evangelicals (see here).

When I received my latest copy of the Wester Recorder, there was an add that took up an entire page of the paper entitled, “Meeting the Calvinist Challenge: A Conference for All Those Seeking Biblical Answers to the Challenge of Calvinism.”  It’s being held at Clear Creek Baptist Bible College on October the 4th from 9:00am to 3:00pm.

Here are the topics that will be addressed:
1) Introduction to Calvinism
2) Calvinism’s Effect on the Public Invitation
3) Does the Gospel of John Teach Unconditional Election?
4) Questions to Ask Prospective Pastors
That last one is the one that scares me most.
Someone else has written a more extensive blog about it here and here.
Someone else (North East Reformation Society) who graduated from Clear Creek Baptist Bible College has given some thought to the conference here.
Calvinism, Arminianism, Calvinism, Arminianism, Calvinism, Arminianism, Calvinists, Arminians, Calvinists, Arminians, TULIP, TULIP, TULIP, TULIP, Calvinism, Arminianism, Calvinism, Arminianism, Calvinism, Arminianism, Calvinism, Arminianism, Calvinism, Arminianism, Calvinism, Arminianism

b l o g t h r i l l

I have begun experimenting with new software for blogging here.  By January of next year, I will no longer be using blogspot.  

THE SHACK ain’t all bad … or … Finally, a balanced review of THE SHACK

My friend Gerald Hiestand has just posted the most refreshing review of the popular book The Shack I have read so far.  

Was Martin Luther Even a Christian? A Critique of His Defense of Infant Baptism

Some might say … based on his defense of infant baptism, that Luther didn’t really believe in the gospel.  

In a previous post where I summarized Luther’s doctrine of baptism, I gave a summary of each of his arguments for infant baptism as presented in the Large Catechism.  This post will take a close look at Luther’s logic for defending infant baptism.  If you have not been following the post series so far, it is important to at least realize this: Luther was a virtual Roman Catholic when it comes to the saving efficacy of the sacraments.  Luther did not believe in the same sola fide that D.A. Carson, R.C. Sproul, and many other modern Reformed evangelicals consider to be fundamental to the gospel.  By R.C. Sproul’s standards, Luther should be considered a non-Christian for not believing in his version of sola fide.  This is an inconvenient truth, for Sproul and many others with his position actually believe they are the modern champions of Luther’s doctrine of justification.  But get this: Luther assumes that baptism saves, period.  As we have seen, he attempts to avoid the accusation of believing in salvation by works by retorting that Baptism is the work of God, and therefore not salvation by works (so argues Luther).  Now we will critique his attempt to establish the validity of infant baptism.  This post will be especially pertinent for baptists. 

Luther’s Defense of Infant Baptism as Involving Logical Fallacies[1] 

                 The Historical Fallacy

Luther’s first argument for infant baptism (based on the fact that infants who are baptized later demonstrate the fruits of the Spirit) commits the historical fallacy known as post hoc, propter hoc (“the mistaken idea that if event B happened after event A, it happened because of event A,”[2]).  His proposal begs an important question: Are the fruits of the Spirit caused by their baptism?  Luther assumes his doctrine of baptismal regeneration to prove his doctrine of infant baptism, but the former, as we have noted, has not been demonstrated on sound principles of interpretation.[3]  His logic could be summarized like this: Since the only way you can possibly receive the Holy Spirit is through baptism (Luther’s assumption), if an infant who is baptized is later shown to bear the fruits of the Spirit, we can safely conclude that their baptism “worked,” and that God blesses infant baptism according to the promise.  Luther assumes the very causal relationship between baptism and salvation which his critics would not be willing to grant.  Surely those critics who opposed Luther on infant baptism were not ready to concede that God saved infants through baptism.  Here we see the very same assumption we have previously observed Luther take for granted in his other arguments against those who deny baptismal regeneration. 

The second argument Luther makes in favor of infant baptism begs the same causal relationship as his first argument.  Luther basically applies the same logic to particular people, namely, the entire history of the church—particularly the early church fathers.  They were baptized in infancy, and we know that God gave them His Holy Spirit.[4]  Therefore, Luther concludes that God has endorsed infant baptism in church history.  His further argument is that God would be in conflict with himself if he were giving the Holy Spirit to people who were baptized as infants if indeed the practice displeased Him.  With this line of argument, Luther continues to assume the causal relationship between infant baptism and the salvation of the church.  I might use this same logic to prove that God obviously blesses sin, since all Christians have sinned by practicing infant baptism and yet God has given them the Holy Spirit.  It may have been (and indeed, I would argue was) in spite of their practice of infant baptism that God gave these men the Holy Spirit.  Furthermore, how would Luther ever be able to demonstrate that anyone who bore the fruits of the Spirit after infancy received the Holy Spirit at their baptism and not some time after their baptism?  Luther seems to be wholly unaware that his arguments will only work for those who share his assumptions about the sacramental limitation of saving grace in baptism.

Begging the Question 

When Luther starts arguing that baptism is valid whether or not faith is present in the one being baptized, it is more difficult to follow his reasoning but easy to see that it is flawed.  When Luther declares that a lack of faith does not “invalidate” baptism because “when the Word accompanies the water, Baptism is valid,” he is continuing to beg the same question he has been begging in every argument we have examined so far.[5]  His further argument that “Baptism does not become invalid even if it is wrongly received or used, for it is bound not to our faith but to the Word,” simply means this: Since God promised to save through baptism without exception, we must assume he saves through baptism without exception.  Again, this begs the important question.  That “we know that God does not lie,”[6] is not enough to prove his case, for only if we assume God promised salvation through baptism could we consider God a liar if baptism did not save an infant.    

Ambiguity and Equivocation 

However, when Luther expands on this idea through illustration, his argument is further complicated by ambiguity.  First, Luther does not make clear whether he is speaking of “validity” from the vantage point of the one who administers the baptism, or from the vantage point of those who receive the baptism.  The illustration of the deceitful Jew who tricks a minister into baptizing him fails to draw the distinction between validity from the vantage point of the dishonest Jew and validity from the vantage point of the administer of his baptism.  The latter would be blameless in the matter.  As far as the rules by which the sacraments are supposed to be administered, he has administered validly.  However, the deceitful Jew has not experienced a valid baptism, because he was not only being baptized for the wrong motives, but without any faith in Christ whatsoever. 

The major problem with Luther’s argument for the “validity” of baptism without faith, therefore, could be described as either ambiguity or as committing the fallacy of equivocation (equivocating the on the term “validity”).  Luther’s illustration is supposed to ground his argument that the sacrament is valid without distinction, but his illustration only proves that baptism can still be administered validly even when it is not received validly.  Because of this, his claim that infant baptism is valid, is either too ambiguous to be proven, or if we assume he means “valid” in an unqualified sense, his illustrations do not help to prove the kind of validity necessary to make a case against his opponents. 

Fallacy of False Analogy 

There is another flaw with the illustrative part of Luther’s defense.  Proving that the baptism can be validly administered to an adult who claims to have faith in Christ but does not (the Jew in Luther’s illustration), is incapable of proving the validity of administering baptism to an adult who does not claim to have faith, much less an infant who cannot even claim to have faith.  If Luther’s illustration cannot prove his claim that baptism should be administered whether or not the recipient has faith, much less is it able to prove the validity of infant baptism.  His argument simply has no particular relevance to infant baptism, in which case the infant is not even claiming to have faith.  Yet this is what Luther is supposed to be proving—that infant baptism is valid because baptism in general remains valid even when “wrongly received or used.”[7]  There is not enough legitimate parallel in Luther’s illustration to carry any weight towards defending infant baptism.  

When Luther attempts to compare the Anabaptist position (that baptism is only valid if faith is present in the recipient) to the position that Christ is only Lord when people believe him to be so, he argues against a straw man.  The differences between these two positions should be painfully obvious, and they consequently nullify Luther’s argument.  Luther’s argument assumes that if his opponents hold that the recipient of baptism must have faith in order for baptism to be valid, we must also grant that things other than baptism are only valid or genuine if someone has faith in them.  Luther is embarrassingly sloppy at this point.  To say that faith is a necessary ingredient for a baptism to be valid (from the perspective of the recipient) is very different than holding that faith is a necessary ingredient for anything to be genuine or valid.  Luther, therefore, although claiming simply to press the logic of his opponents to its absurd conclusion, is in fact pressing the position of his opponents beyond what their position logically demands.  Thus, Luther is guilty of caricaturizing his opponent’s position.  We might return the fallacy by using the same logic against Luther, and so argue that his position (that baptism is valid even without faith) requires him to hold that the whole Christian life might be lived validly apart from any true faith.

Incoherency Within Luther’s Own Position 

An important point to notice about Luther’s defense of infant baptism is the tension which results from his separation of faith from the moment baptism is administered.  On the one hand Luther holds that baptism is not efficacious unless it is “received by faith,” but on the other hand he holds infant baptism to be “valid” even when the infant does not have faith before, during, or years after the baptism.  The necessary inference which must be drawn, then, is that baptism is not always immediately efficacious.  Although, as we have already seen, Luther admonishes his readers to believe (“if you did not believe before [when you were baptized], then believe afterward and confess”[8]), this does not resolve the tension between his holding to the efficacy of baptism on the one hand, and delayed effects of baptism on the other.  We might call this doctrine delayed efficacy, and it is a definite tension which goes unresolved in The Large Catechism.  If Luther holds out this possibility—that faith need not be present at the moment of baptism or even years after—then Luther would be wrong to say that baptism is efficacious in the normal sense of sacramental efficacy.  It would have been more logically clean for Luther to hold that infants receive faith through the baptism rather than holding out the possibility of infant baptism with a delayed efficacy.   

Another tension which is never resolved in Luther’s framework involves his warning about separating faith from baptism.  First of all, Luther himself is guilty of separating faith from baptism with his doctrine of delayed efficacy.  Secondly, his admonishment to others not to separate the two is out of place in his paradigm.  This second problem follows from the fact that real faith cannot exist apart from the regenerative work of the Holy Spirit, who, in Luther’s paradigm, is received only through baptism.  In other words, if faith is a gift of the Holy Spirit, and the Holy Spirit is not received until baptism, then it does not make sense to admonish anyone to “receive” their baptism in faith.  It would make more sense to admonish them to receive their faith in faith, but this does not comport well logically either.

Furthermore, if faith must come through baptism (unless God is a liar who promised the Holy Spirit in baptism, with all his gifts)—that is, if faith cannot not be a result of baptism—what need is there to warn people not to separate faith from baptism?  Given Luther’s paradigm, there is no possible danger of separating faith from baptism, because faith is effectively produced in baptism by God himself.[9]  Yet Luther insists on not administering baptism when faith is not present: “Baptism helps no one and is to be administered to no one unless he believes for himself.  No one who does not personally believe is to be baptized.”[10]  Luther’s admonishments not to separate faith from baptism, then, are not fitting for two reasons: 1) He himself temporally separates the two from one another in infant baptism and 2) his view of efficacy makes such a separation impossible.  For these reasons, it is a logical headache to follow Luther in his incoherent attempt to justify infant baptism apart from faith.[11] 


[1] Luther was forced to use logic, because he admitted that there were no direct commands in the New Testament to baptize infants.  He thought that since the great commission was a simple command to baptize without any mention of faith as a condition, infant baptism was implicated in the command.  Lohse, Martin Luther’s Theology, 303.  Althaus, The Theology of Martin Luther, 361.

[2] Carson, Exegetical Fallacies, 133.

[3] Karl Barth believed that “Luther’s defense of infant baptism is sustainable once the presuppositions of his wider theology of baptism are admitted.”  Trigg, Baptism in the Theology of Martin Luther, 4.  This is precisely because Luther’s defense everywhere assumes his paradigm and begs the key questions.  

[4] It seems that Lohse misses this crucial part of Luther’s argument at this point.  Lohse summarizes this argument like this: “God would not have allowed something improper to be in force for so long.”  Lohse, Martin Luther’s Theology, 303.  However, Luther’s argument was not simply that infant baptism had always been practiced, but that unless God was giving his Holy Spirit through those baptisms, there would be no church.  “…in short, all this time down to the present day no man on earth could have been a Christian.”  Luther, The Book of Concord, 441.  If baptism is the only way God gives his Holy Spirit, and the church members were all baptized in infancy not as adults—the Holy Sprit had to have been given through their baptism or else there would be no church.  Althaus first gives a similar summarization as Lohse, but sees Luther as giving two arguments instead of one.  The second argument he summarizes like this: “It also must be said that if infant baptism were false and contrary to God’s will, then there would have been no true baptism and thus also no church for more than a thousand years.  For without baptism there is no church.”  Althaus, The Theology of Martin Luther, 360.  Luther, however, never voices any such argument in the Catechism, but only speaks in terms of the Holy Spirit’s being present in the church.  Perhaps Althaus is aware of a passage that would shed more light on what Luther meant in the Large Catechism.  Otherwise it seems to be reading into the Luther’s argument.    

[5] Lohse rightly recognizes that Luther appealed to “the concept of the sacrament as ‘effective in itself’ (ex opere operato)” in his defense of infant baptism.  Lohse, Martin Luther’s Theology, 302. 

[6] Luther, The Book of Concord, 444. 

[7] Ibid., 443. 

[8] Ibid., 443. 

[9] “Yet even if they could establish that children are without faith when they are baptized it would make no difference to me … for faith doesn’t exist for the sake of baptism, but baptism for the sake of faith.”  Martin Luther, Luther’s Works, vol 40, Church and Ministry II, 240-41.    

[10] Althaus quoting Luther, The Theology of Martin Luther, 364. 

[11] Althaus says in regard to the problem of making faith necessary and still seeing infant baptism as valid, “Luther’s thoughts about this are not always constant but are in a process of development.”  Ibid., 364.

%d bloggers like this: