Prayer in the Bible

Prayer is one of the most pervasive practices in the biblical texts. The Hebrew Bible contains prayers embedded in narrative, codified in liturgy, and formalized in the Psalter. The New Testament presents prayer as central to the life of Jesus and the early church. The biblical prayer traditions encompass a wide range of genres — lament, petition, intercession, thanksgiving, praise, confession, and contemplative silence — each with its own conventions, theological assumptions, and expected modes of divine response. This article surveys the biblical prayer traditions across both testaments, the theological frameworks used to interpret divine responsiveness, the New Testament’s specific promises about prayer efficacy, and the empirical research on intercessory prayer.^{5, 4}

Prayer in the Hebrew Bible

Balentine identifies prayer in the Hebrew Bible as a form of “divine-human dialogue” in which the human party addresses God with an expectation of response. The Hebrew Bible uses several terms for prayer: tefilah (the general term for prayer), tehinnah (supplication for mercy), rinnah (a cry or shout), and siah (meditation or complaint). The diversity of vocabulary reflects the diversity of prayer forms in Israelite worship.⁵

The patriarchal narratives contain prayers embedded in dialogue. Abraham negotiates with God over the fate of Sodom, bargaining the threshold of righteous inhabitants downward from fifty to ten (Genesis 18:22-33). The negotiation presupposes that God’s intentions can be influenced by human appeal — Abraham addresses God as a judge whose justice is subject to challenge: “Shall not the Judge of all the earth do what is just?” (Genesis 18:25, NRSV). Moses’s intercession after the golden calf episode (Exodus 32:11-14) goes further: the text states that “the LORD changed his mind about the disaster that he planned to bring on his people” (NRSV). Fretheim identifies this pattern as a key feature of the Hebrew Bible’s prayer theology: God is presented as genuinely responsive to human prayer, not merely performing a predetermined response.¹¹

The prophetic tradition complicates this model. Heschel identifies the prophets as figures who stand in a unique relationship to God — simultaneously receiving divine communication and interceding on behalf of the people. Jeremiah’s confessions (Jeremiah 11:18-12:6; Jeremiah 15:10-21; Jeremiah 17:14-18; Jeremiah 18:18-23; Jeremiah 20:7-18) represent a form of prayer that protests God’s actions and questions divine justice. But God also refuses prophetic intercession: “As for you, do not pray for this people, do not raise a cry or prayer on their behalf, and do not intercede with me, for I will not hear you” (Jeremiah 7:16, NRSV). The prohibition against intercession — repeated in Jeremiah 11:14 and Jeremiah 14:11 — presupposes that intercession normally works, otherwise the prohibition would be unnecessary.^{6, 5}

The Psalter as prayer book

The book of Psalms functions as Israel’s prayer book, containing 150 compositions spanning multiple centuries and genres. Broyles categorizes the psalms into major types: hymns of praise (e.g., Psalm 8, Psalm 19, Psalm 104), thanksgiving psalms (e.g., Psalm 30, Psalm 116), individual laments (e.g., Psalm 13, Psalm 22, Psalm 88), communal laments (e.g., Psalm 44, Psalm 74, Psalm 80), royal psalms (e.g., Psalm 2, Psalm 110), and wisdom psalms (e.g., Psalm 1, Psalm 73). Approximately one-third of the psalms are laments — prayers of complaint, protest, and petition arising from suffering.³

Brueggemann identifies a characteristic three-part movement in the psalms of lament: address to God, complaint describing the situation of distress, and petition for divine intervention, often accompanied by a vow of praise conditional on God’s response. Most individual laments move from complaint to confidence — the tone shifts from desperation to trust, sometimes abruptly, as in Psalm 22, which begins “My God, my God, why have you forsaken me?” and ends with universal praise. But Psalm 88 stands as the exception: it ends in darkness without resolution — “You have caused friend and neighbor to shun me; my companions are in darkness” (Psalm 88:18, NRSV). Brueggemann reads the preservation of this psalm in Israel’s liturgy as a deliberate theological choice: the tradition refuses to guarantee that every prayer of lament will be answered with deliverance.^{12, 4}

The psalms of praise and thanksgiving presuppose that God has responded to prior prayer. Psalm 116:1-2 states: “I love the LORD, because he has heard my voice and my supplications. Because he inclined his ear to me, therefore I will call on him as long as I live” (NRSV). Psalm 34:4: “I sought the LORD, and he answered me, and delivered me from all my fears” (NRSV). The thanksgiving genre testifies to experienced divine response, creating a counterpoint to the laments that protest divine silence.³

New Testament prayer promises

The Synoptic Gospels and the Gospel of John contain a series of statements attributed to Jesus that promise divine responsiveness to prayer in sweeping, often unconditional language. Brown notes that the Johannine prayer promises are concentrated in the Farewell Discourse (John 14–16) and are linked to the concept of praying “in my name,” which in Johannine theology signifies alignment with Jesus’s mission and identity rather than a verbal formula.⁷

“Ask, and it will be given you; search, and you will find; knock, and the door will be opened for you. For everyone who asks receives, and everyone who searches finds, and for everyone who knocks, the door will be opened.”

“Whatever you ask for in prayer with faith, you will receive.”

“Therefore I tell you, whatever you ask for in prayer, believe that you have received it, and it will be yours.”

“I will do whatever you ask in my name, so that the Father may be glorified in the Son. If in my name you ask me for anything, I will do it.”

“If you abide in me, and my words abide in you, ask for whatever you wish, and it will be done for you.”

James extends the promise to specific medical contexts: “Are any among you sick? They should call for the elders of the church and have them pray over them, anointing them with oil in the name of the Lord. The prayer of faith will save the sick, and the Lord will raise them up” (James 5:14-15, NRSV). Johnson notes that the passage uses the future indicative (sosei, “will save”) rather than a subjunctive or conditional — the grammar presents healing as the expected outcome of the prayer of faith.⁹

Qualifications and conditions

Other New Testament texts introduce conditions that qualify the sweeping prayer promises. First John states: “And this is the boldness we have in him, that if we ask anything according to his will, he hears us” (1 John 5:14, NRSV). The phrase “according to his will” introduces a condition absent from the Synoptic and Johannine promises — prayer is effective only when it aligns with God’s prior intention. James introduces a different condition: “You ask and do not receive, because you ask wrongly, in order to spend what you get on your pleasures” (James 4:3, NRSV). Paul’s experience adds another dimension: he prayed three times for the removal of a “thorn in the flesh” and received the response “My grace is sufficient for you, for power is made perfect in weakness” (2 Corinthians 12:7-9, NRSV) — a divine refusal rather than an affirmative answer.¹⁵

Jesus’s own prayer in Gethsemane stands as the most theologically significant instance of unanswered petition. In the Synoptic accounts, Jesus prays: “Abba, Father, for you all things are possible; remove this cup from me; yet, not what I want, but what you want” (Mark 14:36, NRSV). Dunn identifies this prayer as paradigmatic: the one who issued the unconditional prayer promises himself receives a negative answer, and the tradition preserves this without harmonization. The Gethsemane prayer introduces a qualifier — “not what I want, but what you want” — that effectively subordinates the petitioner’s request to divine sovereignty.⁸

Luke’s Gospel includes two parables about prayer that emphasize persistence. The parable of the persistent widow (Luke 18:1-8) is introduced with the editorial note that Jesus told it “about their need to pray always and not to lose heart” (NRSV). The parable of the friend at midnight (Luke 11:5-8) similarly commends persistence — the sleeper responds “because of his persistence.” These parables imply that prayer may not be immediately effective and that repeated petition may be necessary, creating a tension with the promises that present divine response as immediate and certain.⁸

The tension in the texts

The biblical texts on prayer contain an internal tension that has generated extensive theological commentary. On one side stand the unconditional promises: “whatever you ask” will be given, “everyone who asks receives,” “the prayer of faith will save the sick.” On the other side stand the qualifications: prayer must be “according to his will,” asked with right motives, offered with faith, and accompanied by persistence — and even Jesus’s own prayer was declined. Balentine identifies this tension as a structural feature of the biblical prayer tradition: the texts affirm both God’s genuine responsiveness to prayer and God’s sovereign freedom to refuse specific requests.⁵

New Testament prayer promises and qualifications^{7, 15}

Theological interpretations

The Christian theological tradition has addressed the tension in the prayer texts through several frameworks. The Augustinian-Calvinist tradition reads the prayer promises as operative within the scope of God’s sovereign will: effective prayer is prayer that aligns with what God has already decreed. Calvin wrote that prayer does not change God’s mind but is the means God has appointed for believers to receive what God has determined to give. On this reading, the qualifications (“according to his will,” “in my name”) are not secondary conditions added to the promises but the interpretive key for understanding them. The promises describe what happens when prayer is properly ordered to divine purposes.⁴

The Arminian-Wesleyan tradition emphasizes genuine divine responsiveness. Prayer is a real factor in divine decision-making — God has sovereignly chosen to make certain outcomes contingent on human petition. The conditional structure (“if you ask, then you will receive”) reflects genuine conditionality, not merely the appearance of conditionality. Fretheim extends this reading into the Hebrew Bible, arguing that texts like Exodus 32:14 (“the LORD changed his mind”) present a God who is genuinely affected by and responsive to human prayer, not a God performing a script.¹¹

The open theism tradition (Clark Pinnock, Gregory Boyd, John Sanders) goes further, arguing that the future is partly open and that prayer can genuinely affect outcomes that God has not predetermined. On this view, unanswered prayer reflects the genuine risk inherent in a world where God has granted meaningful creaturely freedom and where not all outcomes are divinely controlled.¹⁴

Empirical studies of intercessory prayer

The question of whether intercessory prayer produces measurable effects on medical outcomes has been the subject of clinical research since the late twentieth century. The first major study was conducted by Randolph Byrd in 1988 at San Francisco General Hospital. Byrd randomized 393 coronary care unit patients into a prayer group and a control group. The study reported that the prayer group had fewer complications on a severity scoring system, though individual outcome measures showed mixed results, and the methodology was later criticized for endpoint selection and inadequate blinding.¹³

The largest and most methodologically rigorous study is the Study of the Therapeutic Effects of Intercessory Prayer (STEP), published in 2006 by Herbert Benson and colleagues at Harvard Medical School. The study enrolled 1,802 patients scheduled for coronary artery bypass graft surgery at six medical centers. Patients were randomized into three groups: those who were told they might or might not receive intercessory prayer (and did receive it), those who were told they might or might not receive prayer (and did not receive it), and those who were told they would definitely receive prayer. The primary endpoint was any complication within 30 days of surgery. The study found no significant difference in complication rates between those who received prayer and those who did not. The group that knew they were receiving prayer had a slightly higher complication rate (59%) than the group uncertain about receiving prayer (52%), though the study was not designed to test this comparison as a primary hypothesis.¹

A 2009 Cochrane systematic review by Roberts, Ahmed, and Hall examined 10 randomized controlled trials involving 7,646 patients. The review found “no significant difference in recovery from illness” between those receiving intercessory prayer and those who did not. The review noted that most studies had methodological limitations and concluded that there was no clear evidence that intercessory prayer improved health outcomes. The reviewers recommended no further research in this area, stating that the existing evidence was sufficient to conclude that intercessory prayer has no demonstrable therapeutic effect.²

Theological responses to empirical findings

The null results of prayer studies have prompted a range of theological responses. Some theologians argue that the studies are methodologically inappropriate to the subject: prayer is a personal relationship with God, not a mechanical force, and testing it under controlled conditions is a category error analogous to testing whether friendship produces measurable cardiovascular benefits. On this view, the biblical prohibition against “testing” God (Deuteronomy 6:16; Matthew 4:7) implies that God will not perform on demand in experimental settings.¹⁴

Others argue that prayer’s efficacy is real but operates through mechanisms that clinical studies are not designed to detect — spiritual transformation, subjective comfort, community solidarity, or effects on the person praying rather than the person prayed for. Thiselton notes that Paul’s prayer language emphasizes the internal transformation of the one who prays (“we do not know how to pray as we ought, but that very Spirit intercedes with sighs too deep for words,” Romans 8:26, NRSV) rather than the external outcomes of petition.¹⁰

A further theological response holds that the null results are exactly what should be expected if prayer operates “according to his will” rather than as a reliable mechanism for producing requested outcomes. If God responds to prayer with “yes,” “no,” or “wait” according to sovereign wisdom, then the statistical distribution of outcomes among prayed-for and non-prayed-for groups would show no measurable difference, because God’s responses would not correlate with the experimental condition. Critics of this response — including philosophers of religion writing on the problem of evil — have noted that an explanation that is compatible with any possible outcome (positive results confirm prayer’s power; null results confirm God’s sovereignty) is unfalsifiable and therefore lacks explanatory force.¹⁴

Unanswered prayer in the canon

The biblical texts themselves contain narratives of unanswered prayer that complicate the promise tradition. David fasts and prays for seven days for the life of his sick infant son, but the child dies (2 Samuel 12:15-23). Paul prays three times for relief from a physical affliction and is refused (2 Corinthians 12:7-9). Jesus prays in Gethsemane that the cup of suffering be removed, and it is not (Mark 14:36). Habakkuk opens with unanswered prayer: “O LORD, how long shall I cry for help, and you will not listen?” (Habakkuk 1:2, NRSV).⁵

The book of Job presents the most extended treatment of unanswered prayer in the Hebrew Bible. Job’s demand for a hearing before God — “Oh, that I had one to hear me!” (Job 31:35, NRSV) — is eventually answered, but not with the vindication Job seeks. God responds from the whirlwind with questions about the mysteries of creation (Job 38-41) without addressing Job’s suffering. Newsom identifies this as a response that reframes the entire question: rather than answering Job’s petition, God challenges the framework in which the petition was made. The divine response does not explain why Job suffers and does not grant the specific relief requested.¹⁶

The preservation of these narratives alongside the prayer promises creates a canon that contains both the confident assertion that God answers prayer and the sober testimony that God sometimes does not. Brueggemann identifies this juxtaposition as characteristic of the Hebrew Bible’s theology: the canon preserves both testimony (“God answers the faithful”) and counter-testimony (“God is silent when I cry”) without resolving the tension between them.⁴

The Lord’s Prayer

The prayer that Jesus teaches his disciples is preserved in two versions. Matthew’s version (Matthew 6:9-13) is longer and more liturgically developed; Luke’s version (Luke 11:2-4) is shorter. In Matthew, the prayer appears within the Sermon on the Mount, introduced by the instruction: “When you are praying, do not heap up empty phrases as the Gentiles do; for they think that they will be heard because of their many words” (Matthew 6:7, NRSV). The prayer contains six petitions: three directed toward God (hallowed be your name, your kingdom come, your will be done) and three directed toward human need (daily bread, forgiveness, deliverance from evil). Dunn notes that the structure moves from God’s purposes to human needs, suggesting that the prayer’s orientation is theocentric rather than petitionary — the worshipper first aligns with divine purposes before making personal requests.⁸

The petition “your will be done, on earth as it is in heaven” (Matthew 6:10, NRSV) echoes the Gethsemane qualifier. Brown notes the theological significance of this: the model prayer that Jesus teaches already contains the subordination of human desire to divine will that characterizes Jesus’s own prayer in Gethsemane. The prayer does not promise that the petitioner’s wishes will be granted but requests that God’s purposes be accomplished.⁷